Modelling and analysis of daylighting relative to energy consumption of a

heritage category A3 building under energy zone 4 in South Africa

Ochuko Kelvin Overen

Student Number: 2485738

Supervisor: Prof Daniel Irurah

A research report submitted to the Faculty of Engineering and Built Environment,

University of the Witwatersrand in partial fulfilment of the requirement for the degree
of Master of Architecture in Sustainable and Energy Efficient Cities

September 02, 2022

 DECLARATION

I, Ochuko Kelvin Overen, declare that this research report is my own unaided work, except
where acknowledged. The report is submitted for the degree of Masters in Urban Studies (in
the field of Sustainable & Energy Efficient Cities – MUS (SEEC)) at the University of the
Witwatersrand. The research report has not been submitted before for any degree, nor is it
under examination in any other University.

Signature of candidate__________________________________ Date …02 September 2022

i
 ABSTRACT

Restriction of participating buildings in the recently promulgated energy performance

certificate (EPC) regulations to buildings completed after 1999 might have a negative impact
on intended outcomes considering that a significant number of public-use buildings in the
country were completed before 1999. Arising from this concern, this study evaluates
daylighting interventions for a heritage instruction-cum-office building with the objective of
reducing overall energy-use intensity (EUI), related operational energy cost and CO2
emissions. A quantitative study approach based on experimental and simulation research
tools was employed on Livingstone Hall (University of Fort Hare, Alice campus in Eastern
Cape) as the case study building.

Through measurements and simulation, indoor daylighting levels were evaluated against the
300 lux threshold (recommended workplace illuminance levels in South Africa). The data
indicate that on typical clear sky days in summer and winter, average daylight illuminance for
50% of the six sample spaces was below 300 lux, while for typical cloudy-sky days in both
seasons, daylight illuminance for all the spaces was below 300 lux. This implies that, for
cloudy-sky days in both seasons, daylight cannot fully replace electric lighting for office and
academic tasks in the building.

Based on simulation results, for a typical summer clear-sky day, the average daylight
illuminance in spaces 41, 43, 49, 11, 15 and 22 was up to 2 000 lux, while average daylighting
in the same spaces on a typical winter clear-sky day was relatively lower. In addition, the
annual “as-is” cumulative energy-use of the building was 49.02 MWh, and overall EUI was
31 kWh/m2/yr. The annual operational energy cost of the building was R67 108.50, while
receptacle equipment (devices such as computers and other appliances connected to
sockets) energy cost was R32 702.14 and the energy cost of electric lighting was R20 091.86.
Cumulatively, the annual CO2 emission of the building was 25 440 kg (averaging 2 120 kg
/month). Automated on/off and continuous dimming (smooth decrease of electric lighting
capacity with ambient illuminance at a pre-set light level) daylight interventions were
considered in the study. Continuous dimming intervention shows the best reduction in annual
lighting energy-use by 64% and overall annual energy-use by 20%, while overall EUI reduced
by 19%. The resultant overall energy cost saving was R12 420.73 while mitigated CO2
emission was 4 839 kg per annum.

Keywords: Daylighting, energy efficiency, energy-use intensity, heritage building, illuminance,

visual comfort

ii
 DEDICATION

To my late mother; Madam Patience Orubor Ogboriaphe

iii
 ACKNOWLEDGEMENTS

My overwhelming gratitude to:

● My supervisor, Prof. Daniel Irurah, for his guidance throughout the research project. I
sincerely appreciate the learning materials and events you shared with me to bring me
up to speed with the updated materials and practical application of my research.
● Late Dr. Gerald Chungu and other staff members in the School of Architecture and
planning for their comments and contributions that indirectly sharpened the research
project.
● The Fort Hare Institute of Technology (FHIT), and Eskom Tertiary Education Support
Program for their financial support. Also, thanks to Prof Edson L. Meyer, the Director
of FHIT, for encouraging and granting me time and opportunity to further my academic
career in Sustainable Energy Efficient City at the University of the Witwatersrand.
● The Head of Department and staff members of the Physics Department in the
University of Fort Hare for granting permission to use their building as a case study in
this research.
● My immense thanks and appreciation to my wife, Mrs Nozuko Z. Manzi-Overen and
my son, Mr Oghenefe M. Overen, for enduring my absence and for moral support.
Also, to my family in Nigeria for their wise words of encouragement and support.
● Lastly and most importantly, to God almighty for granting the health, wisdom and
supportive people to complete this research project successfully.

iv
 TABLE OF CONTENTS

DECLARATION ........................................................................................................................i
ABSTRACT .............................................................................................................................. ii
DEDICATION .......................................................................................................................... iii
ACKNOWLEDGEMENTS ....................................................................................................... iv
TABLE OF CONTENTS ...........................................................................................................v
LIST OF FIGURES ............................................................................................................... viii
LIST OF TABLES .................................................................................................................... xi
1 Chapter 1 ............................................................................................................................. 1
INTRODUCTION, CONTEXTUALISATION AND MOTIVATION OF THE STUDY................. 1
1.1 Background and Contextualisation of the Study ................................................... 1
1.2 Theoretical Contextualisation ................................................................................ 3
1.2.1 Theoretical overview of daylight. ................................................................... 3
1.2.2 Lighting control and daylighting for energy savings in commercial buildings. 5
1.2.3 Lighting energy efficiency interventions for commercial buildings in South
Africa 6
1.3 Rationale ............................................................................................................... 7
1.4 Problem Statement ............................................................................................... 7
1.5 Aim and Objectives ............................................................................................... 8
1.6 Research Question and Related Sub-questions ................................................... 8
1.7 Working Hypothesis .............................................................................................. 9
1.8 Expected Findings ................................................................................................ 9
1.9 Definition of Key Terms ......................................................................................... 9
1.10 Delineation .......................................................................................................... 10
1.11 Structure of the Research Report ....................................................................... 11
2 Chapter 2 ........................................................................................................................... 13
LITERATURE REVIEW – DAYLIGHT-LINKED CONTROL FOR LIGHTING ENERGY
EFFICIENCY......................................................................................................................... 13
2.1 Introduction ......................................................................................................... 13
2.2 Theoretical overview of Building Orientation in the Northern and Southern
Hemispheres ................................................................................................................. 14
2.3 Daylighting Design .............................................................................................. 16
2.3.1 Design Strategies for Daylight Optimisation ................................................ 18
2.4 Lighting Energy Efficiency Strategies ................................................................. 20
v
 2.4.1 Occupancy-based light control .................................................................... 22
2.4.2 Daylight-linked control ................................................................................. 23
2.5 Estimating Energy Savings of Lighting Interventions .......................................... 25
2.6 Conclusions ........................................................................................................ 27
3 Chapter 3 ........................................................................................................................... 29
METHOD OF THE STUDY - BUILDING DAYLIGHT LEVEL MEASUREMENT, ENERGY
PERFORMANCE ESTIMATION AND SIMULATION ........................................................... 29
3.1 Introduction ......................................................................................................... 29
3.2 Description of the Case Study Building ............................................................... 32
3.2.1 Data needed, collection and analysis approaches ...................................... 33
3.2.2 Overview of the Installed Instrumentation .................................................... 33
3.3 As-is Daylighting of the Case Study Building ...................................................... 35
3.3.1 Experimental approach ................................................................................ 35
3.3.2 Simulation approach .................................................................................... 38
3.4 Lighting Energy Performance of the Building ...................................................... 43
3.4.1 Electric lights and appliances in the building ............................................... 43
3.4.2 Occupancy and operating hours .................................................................. 44
3.4.3 Estimation of lighting energy for Selected offices ........................................ 44
3.4.4 Whole-building energy use intensity simulation ........................................... 46
3.5 Data Collection, Presentation, Analysis, and Discussion .................................... 46
3.6 Ethics Consideration ........................................................................................... 47
4 Chapter 4 ........................................................................................................................... 48
AS-IS DAYLIGHTING AND POTENTIAL ENERGY SAVINGS OF THE BUILDING ............ 48
4.1 Introduction ......................................................................................................... 48
4.2 Experimental-based Daylighting Performance Analysis of the Building.............. 48
4.2.1 Outdoor ambient solar radiation and light level ........................................... 48
4.2.2 Indoor daylight illuminance .......................................................................... 50
4.2.3 As-is daylighting energy performance and potential .................................... 59
4.3 Conclusions ........................................................................................................ 66
5 Chapter 5 ........................................................................................................................... 68
Simulation and Optimisation of the Building Daylight and Lighting Energy Performances ... 68
5.1 Introduction ......................................................................................................... 68
5.2 Simulation-based Analysis of Daylighting and Energy Performance Analysis of the
Building ......................................................................................................................... 68
5.2.1 Sun path and solar distribution for the building ............................................ 69

vi
 5.2.2 Daylight distribution and performance of the building .................................. 74
5.3 Whole-building (as-is) Energy Performance ....................................................... 89
5.4 Lighting Energy Optimisation of the Building: Daylight-linked Interventions ....... 98
5.5 Conclusions ...................................................................................................... 105
6 Chapter 6 ......................................................................................................................... 107
Consolidation of Sub-findings, Conclusions and Recommendations .................................. 107
6.1 Introduction ....................................................................................................... 107
6.2 The “as-is” Energy-use Intensity of the Building ............................................... 107
6.3 The Building Key Daylighting Interventions ...................................................... 108
6.4 “As-is” Daylighting Contribution to the Building Energy-use Intensity ............... 109
6.5 Feasible Daylighting Interventions for Energy-use Reduction .......................... 109
6.6 Consolidation of Findings and Conclusions ...................................................... 110
6.7 Limitations and Recommendations ................................................................... 112
References.......................................................................................................................... 116
Appendix A-1: Inventory of ground floor lights and appliances ........................................... 120
Appendix A-2: Inventory of top floor lights and appliances ................................................. 121
Appendix B-1: Ground floor spaces solar shading .............................................................. 122
Appendix B-2: Top floor spaces solar shading ................................................................... 127
Appendix C-1: Ground floor spaces GVCP and DGI .......................................................... 132
Appendix C-2: Top floor spaces GVCP and DGI ................................................................ 133
Appendix D-1: Ethics clearance .......................................................................................... 134

vii
 LIST OF FIGURES

Figure 1.1: Solar earth-atmosphere energy balance system (Hinrichs and Kleinbach,
2013:156) ................................................................................................................................ 4
Figure 2.1: Revolution of the earth around the sun illustrating the Northern and Southern
Seasons; adapted from (Hinrichs and Kleinbach, 2013:157). ............................................... 14
Figure 2.2: Seasonal sun path at the northern and southern hemispheres; adapted (Szokolay,
2008:164) .............................................................................................................................. 15
Figure 2.3: Site assessment approach for daylighting; (a and b) permissible height indicator
(Szokolay, 2008:164) and (c and d) zone of space of natural light inspection (Boodu, 2021:36)
.............................................................................................................................................. 17
Figure 2.4: Type of fenestration for daylighting design (Bradshaw, 2006:280) ..................... 18
Figure 2.5: Summer and winter seasons’ solar penetration in north oriented building in the
southern hemisphere ............................................................................................................ 20
Figure 2.6: Daylight-linked automatic lighting control: (a) atrium building in Quebec City,
Canada (Atif and Galasiu, 2003:443) and (b) passive solar design office building in Kuala
Lumpur, Malaysia (Lim et al., 2017:169) ............................................................................... 26
Figure 3.1: Workflow of the research process ...................................................................... 30
Figure 3.2: The Physics Building at the University of Fort Hare; (a) front view and (b) rear view
.............................................................................................................................................. 32
Figure 3.3: GIS imagery of the University of Fort Hare, Alice in Eastern Cape (a); imagery of
Livingstone Hare at the University of Fort Hare, and a photo of the Physics Building .......... 33
Figure 3.4: Onset HOBO U12-012 Temp/RH/Light/Ex sensor .............................................. 34
Figure 3.5: Solar radiation and weather station at the SolarWatt Park, University of Fort Hare
.............................................................................................................................................. 34
Figure 3.6: The ground floor layout of the Physics building indicating selected offices. ....... 36
Figure 3.7: The top floor layout of the Physics building indicating selected offices .............. 36
Figure 3.8: Solar weather station at the SolarWatt Park, University of Fort Hare ................. 37
Figure 3.9: The ground floor layout of the Physics building at the University of Fort Hare ... 39
Figure 3.10: The top floor layout of the Physics building at the University of Fort ................ 40
Figure 3.11: The various elevations of the building .............................................................. 41
Figure 3.12: The two main window sizes in the building; (a) small and (b) large .................. 42
Figure 3.13: A 3D SunCast model of the case study building on the 15 July ....................... 43
Figure 3.14: Electronic and magnetic fluorescent tubes used in the building. ...................... 44

viii
Figure 4.1: Measured global horizontal irradiance and outdoor horizontal daylight illuminance
in the monitoring period ........................................................................................................ 49
Figure 4.2: Indoor illuminance of selected offices in the case study building ....................... 50
Figure 4.3: Hallway of the selected ground floor selected offices ......................................... 51
Figure 4.4: A photo of the inner space of space 41 .............................................................. 53
Figure 4.5: Sun rays falling on the desk and light sensor in space 11 .................................. 54
Figure 4.6: Summer season: (a) clear sky day global horizontal irradiance and horizontal
daylight illuminance, (b) clear sky day indoor daylight illuminance of selected offices, (c) cloudy
sky day global horizontal irradiance and horizontal daylight illuminance, and (d) cloudy sky
day indoor daylight illuminance of selected offices ............................................................... 55
Figure 4.7: Winter season: (a) clear sky day global horizontal irradiance and horizontal daylight
illuminance, (b) clear sky day indoor daylight illuminance of selected offices, (c) cloudy sky
day global horizontal irradiance and horizontal daylight illuminance, and (d) cloudy sky day
indoor daylight illuminance of selected offices ...................................................................... 58
Figure 4.8: Typical summer season clear sky daylight energy consumption and daylight energy
savings .................................................................................................................................. 61
Figure 4.9: Typical winter clear sky day lights energy consumption and daylight energy savings
.............................................................................................................................................. 63
Figure 4.10: Summer season clear sky day light energy cost of the selected offices ........... 65
Figure 4.11: Winter season clear sky day light energy cost of the selected offices .............. 66
Figure 5.1: Sun path diagram for the latitude of the building ................................................ 69
Figure 5.2: Simulated hourly solar exposure of the building on a typical summer day ......... 72
Figure 5.3: Simulated hourly solar exposure of the building on a typical winter day ............ 73
Figure 5.4: Simulated typical summer and winter days outdoor ambient luminance for various
orientations ........................................................................................................................... 74
Figure 5.5: Simulated indoor luminance of the ground floor spaces on a typical summer day
.............................................................................................................................................. 76
Figure 5.6: Simulated indoor luminance of the ground floor spaces on a typical winter day 77
Figure 5.7: Simulated indoor luminance of the top floor spaces on a typical summer day ... 80
Figure 5.8: Simulated indoor luminance of the top floor spaces on a typical winter day ...... 81
Figure 5.9: Simulated indoor illuminance of the ground floor spaces on a typical summer day
.............................................................................................................................................. 83
Figure 5.10: Simulated indoor illuminance of the ground floor spaces on a typical winter day
.............................................................................................................................................. 84
Figure 5.11: Simulated indoor illuminance of the top floor spaces on a typical summer day 87

ix
Figure 5.12: Simulated indoor illuminance of the top floor spaces on a typical winter day ... 88
Figure 5.13: Simulated average summer lighting and appliances demand on (a) typical
weekdays and (b) typical weekend ....................................................................................... 92
Figure 5.14: Simulated average winter lighting and appliances demand on (a) typical
weekdays and (b) typical weekend ....................................................................................... 93
Figure 5.15: Simulated monthly energy consumption of the building ................................... 94
Figure 5.16: Simulated building electrical appliances and lights energy usage and overall
energy use intensity .............................................................................................................. 95
Figure 5.17: Simulated monthly energy cost of the building ................................................. 96
Figure 5.18: Simulated total energy cost for appliances and lights ....................................... 96
Figure 5.19: Simulated environmental impact of the building energy consumption .............. 97
Figure 5.20: Automatic and continuous dimming equation graphs ....................................... 98
Figure 5.21: Simulated “as-is” and daylight interventions electric lighting demand on (a)
summer weekdays and (b) summer weekends .................................................................... 99
Figure 5.22: Simulated “as-is” and daylight interventions electric lighting demand on (a) winter
weekdays, and (b) winter weekends ................................................................................... 100
Figure 5.23: The building’s monthly “as-is” versus optimised light energy consumption .... 102
Figure 5.24: The building overall energy-use intensity due to automatic on/off and continuous
dimming daylight-linked control sensors ............................................................................. 103
Figure 5.25: Monthly “as-is” and optimised lighting energy cost of the building ................. 104
Figure 5.26: The building optimised lighting energy environmental mitigation ................... 105
Figure 6.1: Daylighting performance summary of selected spaces on the building’s ground
floor ..................................................................................................................................... 110
Figure 6.2: Daylighting performance summary of selected spaces on the building's top floor
............................................................................................................................................ 111
Figure 6.3: Experimental daylight evaluation approach ...................................................... 113
Figure 6.4: Annual sunlight exposure of the building; (a) top floor and (b) ground floor ..... 114

x
 LIST OF TABLES

Table 1-1: Preliminary structure of the research report. ....................................................... 11

Table 3-1: Summary of data needed, collection tools, and analyses relative to the main
research question ................................................................................................................. 31
Table 3-2: Radiometers specifications (Overen, Meyer and Makaka, 2021) ........................ 38
Table 3-3: Thermophysical properties of glass as adapted from IES-VE building material
properties database .............................................................................................................. 42
Table 4-1: Lighting performance of the selected offices in the case study building .............. 51
Table 4-2: Typical clear and cloudy sky illuminance levels of the selected offices in the summer
season .................................................................................................................................. 56
Table 4-3: Typical clear and cloudy illuminance of the selected office in the winter season 59
Table 4-4: Summer clear sky day lighting energy performance of the selected offices ........ 62
Table 4-5: MEGA FLEX Nightsave Urban Small tariff energy charges (Eskom, 2021) ........ 64
Table 5-1: Annual solar altitude of the building ..................................................................... 71
Table 5-2: Comparison of ground and top floor spaces luminance ...................................... 82
Table 5-3: Statistical description of the building ground floor spaces illuminance ................ 85
Table 5-4: Statistical description of the building’s top floor spaces illuminance .................... 89
Table 5-5: Schedule for energy performance simulation of the building ............................... 90
Table 5-6: Seasonal and annual “as-is” and daylight interventions lighting energy demand of
the building.......................................................................................................................... 101

xi
 1CHAPTER 1

INTRODUCTION, CONTEXTUALISATION AND MOTIVATION OF THE STUDY

1.1 Background and Contextualisation of the Study

According to the International Energy Agency (IEA) and United Nations Environment
Programme (UNEP) (2019), buildings account for approximately 40% of global final energy
consumption and related greenhouse gas emissions. This is mainly due to energy-use in the
operation and construction of residential, commercial, and other buildings in both public and
private sectors. In view of the 7% building consumption increase between 2010 and 2018, IEA
and UNEP (2019) emphasise that even though many countries have set building energy
efficiency initiatives to achieve the UN SDGs (sustainable development goals) and the Paris
Climate Change Agreement targets, building energy reduction was not on track to meet the
set targets.

As highlighted in the World Energy Council (2020:87), various countries have established
action priorities to achieve energy transition to meet global humanity energy challenges. Action
priorities in this context are the economic, energy efficiency and renewable energy activities
embarked on by global energy leaders to achieve energy transition (World Energy Council,
2020). In this regard, Germany set a target to reduce building energy consumption by 20%
and 50% by 2020 and 2050, respectively. China has also adopted electricity price subsidies
to encourage energy efficiency practices (World Energy Council, 2020:111). Switzerland has
a similar approach that incentivises energy efficiency retrofit in buildings (ibid.:119). On the
other hand, South Korea uses digital technology to reduce energy consumption in residential
and commercial buildings through a comprehensive database and automated interventions
coupled with artificial intelligence. (ibid.:45).

Similarly, as a signatory to the Paris Agreement, South Africa has formulated its building
energy efficiency action priority. As of December 2020, it is mandatory by law for all new and
existing public buildings completed since 1999 to display and submit energy performance
certificate (EPC) (Department of Mineral Resources and Energy, 2020). The energy
performance ratings of a building in the EPC programme are based on SANS1544 of the South
African Bureau of Standards - SABS (2021) recommended annual maximum energy
consumption of the various occupancy categories in the different energy zones. Hence,
building energy performance varies with occupancy-type categories and climate zones. The
occupancy-type categories include hospitals, hotels, offices, places of worship, instruction,

1
entertainment and public assembly. Other criteria for EPC requirements are that the building
should have a minimum net floor area of over 2 000 m2, no major renovation within the past
two years of operation and must have been completed after 1999 (Milford, 2015:6).

It is anticipated that the next step in the EPC process will be to set targets/standards in energy-
use intensity (EUI) for existing public buildings (similar to what has been in force for new
buildings since 2011), followed by enforcing compliance through energy efficiency
interventions. In view of daylighting optimisation towards reducing energy for lighting in such
buildings ((Li, Wu, Yuan and Zuo, 2021)), this study envisages that retrofitting of effective
daylighting interventions will become significant in the subsequent stage of optimising energy
performance, especially in the contribution towards the country’s climate change mitigation
commitments as well as towards energy security.

While the rapid innovations and diffusion of lighting technologies, from incandescent and
magnetic ballast fluorescent fittings to light-emitting diode (LED) and electronic ballast
fluorescent fittings, have reduced lighting energy consumption in buildings, daylight remains
the only free, zero-energy and zero-carbon light source (Eskom, 2015, Kwong, 2020 and
Mantzourani, Doulos, Kontadakis and Tsangrassoulis 2020). In addition to its energy savings
potential, studies such as Chinazzo, Wienold and Andersen (2019), Öner, Kazanasmaz,
Leccese, and Salvadori (2020) and Wirz-Justice, Skene and Münch (2020) clearly
demonstrate that daylighting improves well-being, health, and productivity in the workplace.
Whereas the energy and psychological benefits of daylighting have been systematically
studied, design-optimisation for daylighting in buildings, and specifically in public buildings,
remains a major challenge (Kwong, 2020). As a response towards addressing the challenge,
Kwong (2020:1) further notes that the Malaysian Government enforces daylighting
optimisation by making it mandatory through its building policy and regulations.

In terms of regulating and evaluating levels of indoor daylighting, qualitative and quantitative
methods can be adopted (Michael, Gregoriou and Kalogirou, 2018). Qualitative methods focus
on daylight’s contribution to visual comfort or stress based on occupants’ perception. In
contrast, quantitative approaches deal with measurable daylight relative to the building design,
interventions, ambient weather conditions and surrounding features of the site/area in which
the building is located.

The quantitative approach is also used to derive potential savings in energy-use for lighting
due to interventions aimed at optimisation of daylighting through design (ibid.). Over the years,
the three commonly used measures for quantifying daylight levels in buildings are Daylight

2
Autonomy (DA), Daylight Factor (DF) and Useful Daylight Illuminance (UDI) (Bian and Ma,
2017 and Zhang and Lu, 2019). These measures are also applied in building information
modelling (BIM) applications such as IES-VE, DiaLux, DesignBuilder and EnergyPlus to
simulate performance in daylighting and related energy-use performance during the early
design stage of a building or for energy efficiency retrofit for existing buildings (Bernett, Kral
and Dogan, 2021; Elakkad and Ismaeel, 2021).

1.2 Theoretical Contextualisation

The key theoretical themes underlying the scope of the study are the concept of daylight,
workplace light level requirements and daylight metrics, as well as lighting energy efficiency
interventions for commercial/public buildings in South Africa and daylight-linked automated
control systems for commercial buildings

1.2.1 Theoretical overview of daylight.

At the upper levels of the earth’s atmosphere, the solar radiation spectrum consists of 9%
ultraviolet, 50% infrared and 40% visible radiation (Hinrichs and Kleinbach, 2013). However,
only approximately half of that radiation reaches the earth's surface due to reflection,
absorption and scattering as it travels through the atmosphere (ibid.). Although reflection alters
the overall flux density of solar radiation reaching the earth’s surface, scattering affects visible
radiation (Kreider, Curtiss and Rabl, 2009). The solar energy balance reaching the earth after
atmospheric interference is presented in Figure 1.1.

At the earth’s surface, solar radiation is sensed mainly as direct normal and diffused horizontal
radiation. Radiation reaching the earth’s surface without attenuation is referred to as direct
normal radiation, while the fraction of solar radiation after atmospheric attenuation is known
as diffuse radiation. A combination of diffuse and direct radiations is referred to as global
horizontal radiation. Lastly, the reflected solar radiation from the earth's surface (land, water,
vegetation cover, as well as buildings) is called albedo or reflected radiation (Overen, Meyer
and Makaka, 2021b). Studies such as Kunwar, Cetin, Passe, Zhou, and Li (2020) and Kwong
(2020) have demonstrated that diffuse and reflected radiations are conducive to daylighting.
In contrast, Kunwar et al. (2020) highlight that direct solar radiation, it contributes to visual
discomfort primarily due to glare, while also likely to contribute to excessive heat-gain inside
the building once short-wave radiation is converted to long-wave radiation (hence the
greenhouse effect), which we then sense as heat.

3
Figure 1.1: Solar earth-atmosphere energy balance system (Hinrichs and Kleinbach,
2013:156)

Wirz-Justice et al. (2020:2) define daylight as atmospherically attenuated indirect and direct
electromagnetic solar radiation reaching the earth’s surface. The authors (ibid.) further
emphasise that daylight duration and availability depend on atmospheric conditions, season,
and geographic latitude. Boodu (2021) also note that daylight availability in a building is
influenced by the interaction between design (windows position/orientation, size and proximity
of ambient obstructions), local climate and specific weather conditions., In a review of
workplace lighting requirements, Choi, Loftness, and Aziz (2012:167) stress that light quality
is crucial for visual comfort and workplace safety as well as productivity. Several countries
have therefore established recommended light levels for various workplaces in order to ensure
workers’ safety, health, and enhanced productivity. For example, the Chartered Institution of
Building Services Engineers of the United Kingdom recommends 300 lux for moderately easy
visual tasks and 500 lux for tasks that require a high level of accuracy (Yu and Su, 2015 and
Zhang et al., 2019). Similarly, the Illuminating Engineering Society in the United States
recommends 500 lux for paper-based tasks in contrast to 300 lux recommended for computer-
based tasks (Choi et al., 2012 and Yu et al., 2015). Krüger and Dorigo (2008) indicate that the
Brazilian Standard NBR 5413 recommend between 200 lux and 500 lux for teaching, lecturing,
and learning activities. The South African Government has set a 300 lux benchmark for
conference rooms and general office tasks (Department of Labour, 1993). These examples
indicate that lighting-level requirements are task-specific and also vary from one country to

4
another. Szokolay (2004:146) emphasises that an effective lighting requirement needs to
consider the contrast, speed, and level of details of the visual task. Szokolay further argues
that age should also be considered because human visual acuity tends to deteriorate with age.

1.2.2 Lighting control and daylighting for energy savings in commercial buildings
No matter how efficient a lamp is, inadequate control will result in excessive energy
consumption (Eskom, 2015). Similarly, the integration of daylight in a building does not
automatically contribute to energy savings. Daylight-linked lighting control is therefore often
required to achieve the desired energy savings. Occupancy and daylight-linked sensors are
the commonly used automated lighting control technologies (Roisin, Bodart, Deneyer and
D’Herdt, 2008 and Yu et al., 2015). Occupancy control operates in two distinctive detection
modes, which are presence and absence. In presence-detection, the sensor automatically
switches on electric lights once motion is sensed and off when the space is not in
use/occupied. Electric lights would still need to be manually switched on, and the sensor then
switches off the light when the occupant vacates the space without doing so (ibid.). Roisin et
al. (2008: 522) assessed the lighting energy-savings potential in office spaces using daylight-
linked, occupancy dimming and on/off lighting controls. Their study concluded that energy
savings due to occupancy lighting control depend on occupation rate and control mechanism.

Furthermore, daylight-linked control relies on photosensors to control the operation of electric

lights based on ambient daylight levels (Overen and Meyer, 2019). Yu et al. (2015) identify
two common daylight-linked controls, which are dimming control and on/off control. Daylight-
linked dimming control sensors use the available daylight level to vary the output of electric
lights (ibid.). In this control, the photosensor is usually mounted on the ceiling and facing
downward towards a specific space (working plane) indoors (Lim, Keumala and Ghafar, 2017).

Yu et al. (2015) further indicate that the electric light output in daylight-linked dimming control
is proportional to energy consumption. Lighting energy savings would therefore be
proportional to levels of daylight transmitted indoors through the windows. On the other hand,
on/off daylight-linked control uses an outward-facing sensor to switch electric lights on or off
based on pre-set illuminance (Lim et al., 2017). In other words, when the indoor ambient
daylight level is equal to or above the pre-set illuminance, the sensor switches off the electric
lights and subsequently on when the indoor daylight level is below the set level (Yu et al.,
2015). Under this control, the total period the indoor daylight level is equal to or above the pre-
set illuminance is used to determine the lighting energy savings (ibid.).

5
Based on the field measurement of an office building, Onaygil and Güler (2003:977) reported
that daylight-linked lighting control saves approximately 30% of lighting energy consumption
under Istanbul’s (Turkey) climate conditions. The authors (ibid.:977) indicate that daylight-
linked energy savings are seasonal, with savings increasing in the summer season by up to
10%. The sky condition is also a contributing factor to the performance of daylight-linked
technology (ibid.). Findings by Alrubaih, Zain, Alghoul, Ibrahim, Shameri, and Elayeb (2013)
corroborate Onaygil and Güler (2003) findings and further suggest that irrespective of the
daylight-linked control strategy, daylighting has the potential for more than 50% lighting
energy-savings. Shailesh and Raikar (2010:33) also argue that about 70% savings on electric
lighting energy consumption could be achieved by combining occupancy and daylight-linked
sensors in a building.

1.2.3 Lighting energy efficiency interventions for commercial buildings in South

Africa
that the invested in nationwide South Africa’s national power utility (Eskom) launched the
national EEDSM in 2004 under the administration of the National Energy Regulator of South
Africa (NERSA) (Etzinger, 2014:1). Etzinger (2014:3) further indicates that in response to the
evolving energy landscape, the scope and focus of EEDSM were changed and renamed
Integrated Demand Management (IDM) which was aimed at promoting energy efficiency
practice in South Africa, specifically in the commercial, residential and industrial sectors
through the use of rebates, standard product, standard offer, energy services company
(ESCo) funding process, and performance contract funding models. Retrofit initiatives such
as energy-efficient lighting, water heating, efficient motors, and heating, ventilation air
condition (HVAC) systems were notable IDM projects in the commercial-building sector
(Eskom, 2013).

IDM also entailed the replacement of inefficient lights such as incandescent and halogen
lamps or fluorescent tubes with magnetic ballast with efficient lights such as LED, compact
fluorescent lamp (CFL) or fluorescent electronic ballast. Motion sensor light controls (on/off)
were also used in some projects. However, overall lighting energy interventions depended on
the nature of the building owner/user organisation, type of task, targeted energy savings as
well as the financial capability of the client or ESCo involved. The key principle of IDM
initiatives target energy savings without compromising workplace safety and productivity at a
time when supportive policies and regulations for existing buildings have not yet been
promulgated (Etzinger, 2014).

6
1.3 Rationale

The prevailing carbon-intensity and electricity-supply crisis facing the South African economy
places an enormous burden on the government in terms of its international climate-change
mitigation commitments while also ensuring energy security for the country. Decoupling
energy consumption and carbon intensity from economic growth through initiatives such as
the newly launched EPC is a substantial step towards reducing strain on the national grid and
the associated greenhouse gas (GHG) emissions. In line with the national building energy
efficiency action priority, modelling and simulation of the energy performance of a heritage
instruction-cum-office building would reveal the extent to which such buildings could also
contribute to the national energy efficiency scheme while also improving on well-being and
productivity of their users. Daylighting analysis of the building covered in the report also
highlights the potential of adopting daylighting interventions to reduce EUI, operational costs
and mitigation of GHG emissions for heritage buildings in the country without major
modifications, which could undermine their heritage status.

1.4 Problem Statement

South Africa is currently undergoing its worst electricity-supply crisis, with persistent load
shedding over the past three years. This is attributed to overstrained national grid due to the
increasing energy demand and ageing coal-fired power plants with inadequate addition of new
generation capacity. The Chief Executive Officer of Eskom (the national electricity utility) has
indicated a high possibility that the ongoing electricity crisis will continue for another five years
(Ash, 2021). Amid this crisis, the energy consumption of commercial buildings increased by
9% between 2015 and 2019 (Eskom, 2019). In 2019, Eskom also recorded its highest
particulate matter and GHG emissions in 20 years (ibid.). The growing energy demand and
ageing power plants also exacerbate the utility’s non-compliance with the country’s
environmental standards and regulations, especially in relation to air-quality impacts of
pollution from coal-power stations.

In addition to the electricity supply shortage and negative environmental impact, Eskom has
persistently sought higher-than-inflation tariff increases in order to mitigate its deteriorating
financial performance, thus indirectly passing the financial burden to electricity consumers,
which in turn undermines economic growth. The national energy efficiency strategy for public
buildings (which now includes EPC regulations) can therefore be viewed as a long term
economic and environmentally friendly approach towards addressing the country’s electricity
challenges. However, reduced participation due to building age (exclusion of buildings

7
completed prior to 1999) will negatively impact the efficacy of the policy, mainly because most
of the country’s stock of public-use buildings were completed before 1999, even when they
are not necessarily heritage buildings (Milford, 2015:6).

1.5 Aim and Objectives

In view of the challenges outlined above and based on a case study of a heritage instruction-
cum-office building, this study evaluates the extent to which daylight optimisation interventions
can reduce overall EUI and operational costs while also contributing to the mitigation of
environmental impacts such as GHG emissions. In order to achieve this aim, the study was
guided by the following objectives:

• Determine status quo daylighting levels by installing photometric sensors and data
loggers in the case study building
• Assess the outdoor luminance levels based on outdoor photometric and solar
radiation sensors and data loggers
• Conduct a walk-through audit of the building envelope components (including
critical dimensions for modelling), materials properties, floor area, count of electric
lights and appliances (including their power ratings), as well as the schedule of use
of the building. These were the critical simulation inputs used to derive the EUI
• Derive overall “as-is” EUI performance of the building as well as daylighting
optimised performance based on a range of feasible interventions and resultant
financial, and environmental benefits

1.6 Research Question and Related Sub-questions

The overall research question which guided the study was articulated as follows:

What is the potential contribution of daylighting optimisation towards EUI-reduction in a

heritage place-of-instruction building in South Africa, and what are the possible daylighting
interventions towards reducing EUI and enhanced financial and environmental benefits of
such buildings?

The sub-questions towards addressing the main research question were articulated around
the case-study building, which falls under energy zone 4 as per the 2nd edition of SANS10400-
XA. The sub-questions were articulated as follows:

• What is the status quo (“as-is”) EUI for the selected case study building?

8
 • What are the potential daylighting optimisation interventions of the case study
building while also adhering to its heritage status?
• What would be the contribution of daylighting optimisation interventions towards the
reduction of EUI of the building?
• What would be the feasibility of retrofitting such daylighting optimisation
interventions?

1.7 Working Hypothesis

Given that climate zone 4 under SANS10400-XA is characterised by 1 765.6 kWh/m2 solar
radiation per annum (SOLARGIS, 2017), an average of 13 hours of daylight in summer and
10 hours in winter season (Overen, Meyer and Makaka, 2021a), daylighting indoors can
enhance EUI-reduction by up to 50% of lighting energy-use as observed in Alrubaih, Zain,
Alghoul, Ibrahim, Shameri and Elayeb, (2013), and thus contribute to global climate change
mitigation. Typical heritage buildings are most likely to be deficient in their design towards
utilising daylighting for EUI-reduction. Heritage buildings in South Africa are also not
commonly considered in energy efficiency optimisation through retrofitting of responsive
interventions as they are normally deemed to be disruptive to the heritage value of such
buildings (Milford, 2015). However, considering the available solar radiation in energy zone 4
and the use of non-invasive lighting interventions such as daylight-linked and occupancy
control, the heritage category A3 building could save up to 50% of lighting energy usage as
well as contribute to operational-cost reduction and mitigation of GHG-emissions.

1.8 Expected Findings

As-is lighting energy and potential daylighting energy saving of selected offices in the building
were anticipated at the end of the research. This involves daylight illuminance of selected
offices at desk level on both floors and a simulated whole-building daylight distribution, building
electrical energy savings due to daylight, resultant cost savings and environmental mitigation.
Based on the findings, lighting energy cost and environmental impact of the building were
deduced. In addition, the whole-building EUI, operational cost, and environmental impact were
simulated. The simulation also covers non-invasive lighting interventions for optimised EUI.

1.9 Definition of Key Terms

• Area of interest (AOI) is the region in a space, usually 0.5 m away from the perimeter
walls and windows, used to evaluate daylight level or quality.

9
 • Automated on/off daylight-linked operation uses a photosensor to automatically
switch on and off of electric lights based on pre-set indoor ambient light levels (Atif
and Galasiu, 2003 and Lim et al., 2017).
• Conditioned spaces are areas or rooms in a building that are mechanically or
electrically thermally conditioned for thermal comfort.
• Continuous dimming daylight-linked operation uses a photosensor to gradually
decrease the capacity of electric lighting with indoor ambient illuminance to a pre-
set light level (Atif and Galasiu, 2003 and Yu et al., 2015).
• Daylight Autonomy is the percentage of annual daytime hours that a work plane is
above or within the recommended light level (Bian et al., 2017).
• Daylight factor is expressed as the ratio of indoor daylight level to unobstructed
outdoor ambient horizontal daylight illuminance under the International Commission
for Illuminance (CIE) Overcast Sky Standard (Tregenza and Mardaljevic, 2018).
• Energy use intensity (EUI) (kWh/m2) is the annual energy consumption of a building
per net total floor area where the floor area only takes into consideration the active
area of a building, thus excluding spaces such as the garage, bathrooms and
outdoor areas.
• Luminance (cd/m2) is the amount of light passing through an opening, emitting, or
reflecting off an object.
• Illuminance (lux) or light level is the amount of daylight falling onto a specific surface
area.
• Side-lighting is a daylighting design strategy that harvests natural lights through
vertical windows, for example, windows on the walls (Alrubaih et al., 2013 and
Bradshaw, 2006).
• Top-lighting is a daylighting design strategy that harvests natural lights through
vertical windows, for example, skylights (Alrubaih et al., 2013 and Bradshaw, 2006).
• Useful daylight illuminance is the annual percentage of daytime hours that a work
plane illuminance is within a specified range of light level (Zhang et al., 2019).
• Work plane is the height at which daylight illuminance in a given space is evaluated,
and it is usually at 0.77 m to 0.85 m above floor level, equivalent to the height of a
typical office desk.

1.10 Delineation

In this study, daylight observation of the building was conducted using experimental and
simulation approaches. In the experimental approach, photometric sensors were used to
10
monitor the illuminance on the desk of selected offices in the building. The measurement
captured the operating daylight conditions of the selected offices, i.e., the effect of the blinds
on the available daylight, electric and ambient lights were involved. Therefore, the daylight
illuminance of the spaces was obtained by deducting the average illuminance during the sun’s
absence from the average illuminance during the sun’s presence. The sun absence and
presence were observed from the measured global horizontal irradiance (GHI) based on the
assumptions that periods with GHI greater than 0 W/m2 imply the sun presence while the sun
absence is the periods with GHI below 0 W/m2. GHI was also used to determine clear and
cloudy sky days for daily daylight evaluation of the building. Further, spaces with daylight
illuminance greater than 300 lux were considered sufficient to utilise daylight for energy
savings.

The simulation approach evaluates the whole-building daylighting and energy performance,
focusing on conditioned spaces. However, daylighting simulation of the building was based
on Bisho town satellite-based weather data and the original building design, which did not
include the windows’ cover and electric lights’ illuminance. Regarding the building energy
performance, lighting and appliance energy forms the whole-building consumption. Their
respective operating periods, as used to simulate the energy consumption, were obtained from
the academic schedules in the building. In addition, the research only covers fixed
fenestration, i.e., allowable fenestration changes for heritage buildings towards improving
daylight indoors were not considered in the research.

1.11 Structure of the Research Report

This section provides the structure of the research report in terms of chapters and content, as
summarised in Table1-1.

Table 1-1: Structure of the research report.

Chapter Heading (title) Description
In this chapter, the research background, problem
1 Introduction
statement, aim and objectives are outlined and discussed
Sample studies on two key themes that the research is
based on are presented and appraised in this chapter.
Literature review (building
The chapter starts with an overview of lighting energy
2 lighting energy efficiency and
efficiency interventions in public (office) buildings.
daylight design)
Thereafter, studies on building design interventions for
optimum daylight performance are outlined and appraised
Research method (daylight The various methods and data tools used in measuring
level and energy the “as-is” daylight level in the offices are presented and
3
measurement, estimation, discussed in this chapter. Assumptions and formulae
and simulation) used in estimating the lighting energy consumption,

11
 operational cost, and environmental impact are presented
and reviewed. The chapter also covers “as-is” daylight
levels and energy performance modelling procedures for
the entire building
Results and discussion (“as-
This chapter presents data and analyses for the “as-is”
is” daylighting and potential
4 daylighting levels of selected spaces in the building and
energy savings of simulated
the resultant energy consumption and operational cost.
interventions)
The chapter presents the simulated whole-building
Results and discussion daylight level, EUI, operational cost, and related GHG
(Simulation and Optimisation emissions. In addition, data and analyses on the various
5 of the Building Daylight and lighting energy interventions adopted to optimise building
Lighting Energy energy performance are presented and discussed in this
Performances) chapter. The discussion covers the optimised EUI,
operational cost and environmental impact of the building.
The chapter presents the key findings of the study in
relation to the objectives and overall research question of
Overall findings, conclusions
6 the study. The chapter also presents the
and recommendations
recommendations for future studies on daylighting in
heritage category A3 buildings in South Africa.

12
 2CHAPTER 2

LITERATURE REVIEW – DAYLIGHT-LINKED CONTROL FOR LIGHTING ENERGY

EFFICIENCY

2.1 Introduction

Globally, building energy consumption has been a central demand-side focus towards
achieving net-zero carbon emission and based on design interventions in lighting, heating,
ventilation and air conditioning (HVAC) and appliances. The International Energy Agency
(IEA) (2021:146) anticipates that electrical appliances and lighting will become much more
efficient in the next three decades leading to 2050 due to energy-use policies as well as
ongoing escalations in technological innovations. As an example, the report envisages that
the proportion of LED lamps in total light fixture sales will reach 100% by 2025 (ibid.). In
advancing the global net-zero energy building target, Hong Kong Green Building Council
launched a green building best practice guidebook (Hong Kong Green Building Council
Limited, 2021), which recommends 32 strategies for smart green buildings under six key
themes: building design and operation, health and well-being, energy performance, material
and waste management, water performance, and mobility and transportation. The health and
well-being theme deals with the use of smart technologies to ensure and maintain building
environmental quality pertaining to occupants’ health and well-being, smart artificial lighting,
advanced solar technologies for natural lighting and smart light poles. The guidebook also
emphasised that the use of advanced solar technologies could save 15 to 40% energy by
reducing energy for artificial light while also enhancing occupant comfort and well-being
(ibid.:36). The Green Building Council South Africa is equivalent to Hong Kong Green Building
Council that promotes energy and environmentally sustainable building design through
certification.

Krarti, Erickson and Hillman (2005) suggest that the lack of evaluation methods for daylighting
suitability and potential energy savings has been the main reason for the ongoing reluctance
towards optimisation of daylighting in buildings. Although numerous studies such as Lee, Lam
and Lee (2018) emphasise the substantial lighting energy-savings through energy efficient
light retrofitting and automation control systems, such studies also stress that the lack of
capital and technological awareness are the main factors hindering building owners from
considering and implementing retrofitting for daylighting optimisation interventions.

13
Guided by this insight, this Chapter reviews studies on the various concepts of lighting energy
efficiency in buildings and basic daylighting design. In particular, the studies reviewed cover
lighting energy efficiency strategies and the application of the two major lighting automation
control systems, which are occupancy-based and daylight-linked. Studies in lighting energy
savings estimation are subsequently reviewed, followed by a discussion on strategies of
daylighting design and a conclusion on the key insights for the study. Some of the key studies
appraised are Lee et al. (2018), Booysen, Samuels and Grobbelaar (2021), de Bakker, Aarts,
Kort and Rosemann (2018), Lim, Keumala and Ghafar (2017), Chi, Moreno and Navarro
(2018), Yu et al. (2015), Szokolay (2008), Boodu (2021), and Bradshaw (2006).

2.2 Theoretical overview of Building Orientation in the Northern and Southern

Hemispheres

After atmospheric attenuation of solar radiation as discussed in Section 1.2.1, the radiation
intensity reaching a given location on the earth’s surface is determined by the latitude, season,
time of day, and degree of cloud cover (Hinrichs and Kleinbach, 2013). According to Hinrichs
and Kleinbach (2013), latitude and season are due to the earth’s elliptical orbit around the sun
as shown in Figure 2.1.

Figure 2.1: Revolution of the earth around the sun illustrating the Northern and Southern
Seasons; adapted from (Hinrichs and Kleinbach, 2013:157).

14
As shown in Figure 2.1, the earth’s spinning axis is tilted about its plane of motion at an angle
of 23.5°. As a result, the south pole is tilted toward the sun during the southern hemisphere’s
summer and away from the sun during winter (Hinrichs and Kleinbach, 2013). Hence, locations
in the southern hemisphere are exposed to higher solar intensity and sunlight hours in the
summer season with a peak on 22 December which is the southern summer solstice. The
reverse scenario occurs in the northern hemisphere, where the north pole is closer to the sun
during the summer season, and peak solar intensity and sunlight hours are experienced on
22 June, the northern summer solstice (ibid.). In the winter season in both hemispheres, solar
radiation is subjected to higher atmospheric absorption and scattering and thus spreads over
a large surface. Consequently, low solar intensity and sunlight hours are experienced at the
earth’s surface during the winter season. Sun paths in the major seasons in both hemispheres
are shown in Figure 2.2.

Figure 2.2: Seasonal sun path at the northern and southern hemispheres; adapted
(Szokolay, 2008:164)

The sun angle above the horizon is referred to as altitude, while azimuth is the sun angle along
the horizontal plane from 0 to 360 (Szokolay, 2008:164). The sun’s altitude and azimuth are
the most crucial parameters used in building design for effective solar access relative to the

15
hemisphere, local climate conditions and building purposes. In this regard, changes in solar
altitude influence seasonal solar penetration, while changes in azimuth angle influence the
hourly solar penetration. In building design, both parameters are used to assess solar access
and need for solar shading in relation to space heating and daylighting (Overen, Meyer and
Makaka, 2021).

2.3 Daylighting Design

Natural light or outdoor light consists of sunlight, which is un-attenuated solar radiation from
the sun, while daylight mainly consists of indirect solar radiation diffused within/through the
atmosphere. Sunlight or daylight availability outdoors is therefore a factor of sky conditions
which could range from clear, overcast and intermediate skies (Szokolay, 2008:149).
According to Al-Obaidi, Ismail and Abdul Rahman (2015:9), the illuminance of clear sky
outdoor daylight is between 10 000 lux to 22 000 lux, while during overcast sky, daylight
illuminance ranges from 5 000 lux to 54 000 lux. On a clear sky day, direct sunlight can fall
between 50 000 lux to 100 000 lux (Al-Obaidi et al., 2015:9). However, Szokolay (2008:151)
stresses that penetrating sunlight (direct radiation) through fenestration may be desirable or
undesirable depending on the task indoors, mainly because such radiation entails high levels
of thermal energy (infrared radiation) that may be desirable in the heating season (winter) or
may result in thermal discomfort in the cooling season (summer) (Chi et al., 2018 and
Szokolay, 2008). Fenestration design is therefore a factor of building lighting requirements,
but increased reliance on daylighting to achieve energy efficiency is likely to result in larger
windows which may pose visual discomfort due to glare and thermal discomfort as well
(Szokolay, 2008).

In addition to weather conditions, obstacles such as nearby buildings constitute another factor
that influences access to daylighting indoors. Inappropriate urban planning may result in
overshading from surrounding buildings (Boodu, 2021). According to Szokolay (2008:152),
the “right to sunlight” was launched in the 19th century but only became a common practice
in 1973 after the global energy crisis. This right created the foundation for many building codes
that provide design requirements for daylighting and procedures for site inspection for potential
daylight blockage before construction (Szokolay, 2004). Some of the approaches used for
daylighting site assessment in relation to obstacles are presented in Figures 2.3 (a) to (d).

16
Figure 2.3: Site assessment approach for daylighting; (a and b) permissible height indicator
(Szokolay, 2008:164) and (c and d) zone of space of natural light inspection (Boodu,
2021:36)

In Figure 2.3 (a) and (b), Szokolay (2008) indicates that the existing building or block from
both ends to points A and B, respectively, receives adequate daylight from both sides of the
planned building. Point C, which is halfway between points A and B, is the most crucial
(Szokolay, 2008). Szokolay (2008:164) indicate that the permissive indicator D shows a
permissible height of 13 m between the sill level of the ground floor at point C and X. From
Figure 1 (c) and (d), Boodu (2021:35) alludes that according to SANS 10400-0: 4.2.1.2.6 and
4.2.1.2.7, the recommended length of zones of space of natural light for habitable rooms in
dwelling houses, residential and commercial buildings is 1/3H where H represent the obstacle
(building) height above the level of the opening head concerned. As defined by the author
(ibid.:32), a zone of space is a volume of space around an opening in an external wall of a
building.

As highlighted above, blockage of ambient sunrays by obstacles such as trees and adjacent
buildings influences access to daylight in a building. However, as indicated in the appraised
studies, building codes and regulations have been established to mitigate such daylight
blockage. The appraised studies, therefore, serve as theoretical principles for evaluating the
poor-lit sections of the case study building due to surrounding trees and adjacent buildings.

17
2.3.1 Design Strategies for Daylight Optimisation
As shown in Figure 2.4, the fundamental daylight design strategies are side-lighting and top-
lighting (Alrubaih et al., 2013 and Bradshaw, 2006). Bradshaw (2006:279) further highlights
that fenestration for daylighting (primarily windows) can also serve other purposes but should
be distinguished from openings for ventilation, passive solar heating and views.

Figure 2.4: Type of fenestration for daylighting design (Bradshaw, 2006:280)

Alrubaih et al. (2013) emphasise that a side-lighting design or vertical window admits daylight
at the wall of a building, while a top-lighting or roof window admits daylight at the upper part
above the ceiling. Bradshaw (2006) argues that top-lighting or roof windows are more effective
for daylighting than side-lighting since light from the sky is of greater quality than that of
reflected landscape, and it is less diminished by surrounding obstacles. Alrubaih et al. (2013)
share the same view, and further indicate that side-lighting is directional and diminish as the
distance from the window or opening increases. Side-lighting is also associated with glares as
the openings fall within occupants’ field of view. In addition, top-lighting also allows designers
to determine the geometry and orientation of daylight openings based on occupants’ desire
without the hindrance of the wall orientation.

In terms of thermal impact, Bradshaw (2006) discourages top-lighting in climates with high
cooling load and also stresses that it is not a suitable solar heat collector for passive heating.
Instead, the author (ibid.) recommends weather-tight, north-facing clerestory windows in the
Southern hemisphere (south-facing in the northern hemisphere) for optimum solar heat gain
in the heating season. Such design should also be accompanied by appropriate horizontal

18
sunshading to avoid overheating in the cooling season. Where heating is not required, south-
facing clerestory windows in the southern hemisphere are appropriate for more uniformly
distributed daylighting (ibid.).

Bradshaw (2006) further highlights that the traditional use of windows (side-lighting) is not the
only approach for effective daylighting. The study recommends that dedicated windows should
be created at the building design stage in order to ensure that other purposes of the window,
such as natural cooling and views, are not interfered with.

However, many heritage buildings, such as the case study building in this study, do not have
dedicated windows for daylighting. They are confined to the use of side-lighting vertical
windows, which also serve as openings for ventilation and views. Another crucial design
strategy covered in the appraised studies is the use of geographical orientation of windows to
selectively admit direct and diffuse components of solar radiation. For example, north-facing
windows in the southern hemisphere will result in a significant amount of solar heat gain and
possible glare in the winter season due to direct solar radiation. In contrast, south-facing
windows in the southern hemisphere can be used to admit diffuse solar radiation with minimal
solar heat gain and are therefore conducive to daylighting optimisation. As shown in Figure
2.2, north-facing facades in the southern hemisphere will receive varying levels of solar
radiation between winter and summer seasons while south-facing facades will be in the shade
throughout the year. In Figure 2.5, direct solar radiation in summer is prevented by the
overhanging roof thus reducing solar heat gain and creating a mild daylight-lit space indoors.

On the other hand, low-angle sun rays in winter penetrate the window into the building,
creating solar heating. The sun rays can also result in glare and visual discomfort depending
on occupant’s position relative to the sun rays and task of the occupant. Sunscreens and other
window covers are usually used to mitigate visual discomfort due to direct or excessive solar
radiation. Modern buildings adopt photo-automated sensors that operate such screens based
on outdoor daylight illuminance. However, this drawback is not experienced in the south-facing
sections of the building that are in the shade all year round. Spatial layout based on lighting
requirements of tasks is another approach to effectively utilise high daylight illuminance on
north-facing zones of a building.

19
Figure 2.5: Summer and winter seasons’ solar penetration in north oriented building in the
southern hemisphere

The principles distilled from the appraised studies on daylighting design strategies served as
crucial guides in analysing the ability of the case study building to admit effective daylight
relative to its design. The studies also served as a theoretical evaluation of sections of the
building with excess or poor light levels due to orientation of windows. In general, the studies
provided a background understanding of daylight distribution relative to the opening type and
geographical orientation of the building. This was then applied towards analysing the
experimental (measured) and simulated data outputs.

2.4 Lighting Energy Efficiency Strategies

The replacement of energy inefficient lights such as incandescent and halogen lamps with
efficient compact fluorescent lamps (CFL) and light-emitting diode (LED) is the most common
lighting energy efficiency intervention for existing buildings. Krarti et al. (2005:747) allude that
using CFL, occupancy-based light control sensors and designs that strategically minimise the
required number of light fixtures can reduce operational energy and cost for new and existing
buildings. Lee et al. (2018) share similar views with Krarti et al. (2005) but go further to argue
that there is a high potential of performance risk in energy performance contracting projects
for energy service companies (ESCos) in relation to lighting retrofit. The authors (ibid.) argue
that the performance of energy efficient lighting retrofit is mainly influenced by factors that go
beyond the ESCos control. The factors include lighting usage pattern, quantity and quality and

20
frequency of maintenance. The study therefore explores a probabilistic approach to evaluate
the performance risks of lighting retrofit interventions such as inefficient lights replacement,
daylight-linked, and occupancy-based light control systems. In the study, EnergyPlus and
Monte Carlo simulation techniques were adopted, while hypothetical T8 fluorescent tubes
were replaced with T5 fluorescent tubes. One of the key study findings was that the potential
annual lighting energy saving of a hypothetical 40 storey office building in Hong Kong would
be between 1 267 and 1 927 MWh. Based on the wide range of energy savings, the study
concludes that lighting retrofit poses performance and investment risk for ESCos engaged in
performance contracting.

In contrast, Booysen, Samuels and Grobbelaar (2021) argue that lighting retrofit is one of the
key approaches for schools in South Africa to navigate the electricity crisis associated with
increasing tariff charges in the country. According to the study, in view of the diminishing costs
of LED lamps, replacing fluorescent fixtures with LED lamps will save utility costs that can be
allocated for much-needed academic services and also, benefit the struggling national utility
and for mitigating environmental concerns such as GHG emissions. The study was based on
energy savings, life-cycle cost, and CO2 emission mitigation by replacing 1 170 58W – 60W
T8 fluorescent fixtures with 19W – 32W LED lamps in sixteen schools in the Western Cape
province of South Africa where the study was conducted. The energy savings per school day
were estimated at 80 to 190 kWh with 80 to 180 kg CO2 emission reduction, while the life-
cycle cost savings vary from R298 180.00 to R766 570.00 based on the quality of lamps and
tariff structures (ibid.:9).

In both studies, the replacement of inefficient lamps as a lighting energy efficiency strategy
has been adequately demonstrated. However, Lee et al. (2018) argue that the associated
performance risk for ESCos may serve as a hindrance for performance contracting based on
light retrofitting. The study further indicates that agreed estimated energy savings can avert
this challenge prior to project implementation. The use of energy efficient lighting can be used
to reduce high utility costs and lessen pressure on the national grid and the environment in
countries such as South Africa, which also has a carbon-intensive energy generation system
(Booysen et al., 2021). In terms of complementary or additional interventions, Doulos,
Tsangrassoulis and Topalis (2008) emphasise that automated light control interventions such
as occupancy-based, timer, and daylight-linked controls can improve lighting energy
performance.

The studies covered in this section are relevant to this research as they reaffirm one of the
simplest approaches to achieving lighting energy efficiency. Some of the interventions, such

21
as replacing inefficient lamps with efficient ones, offer lighting energy savings intervention
options evaluated in this study. CO2 emission-reduction assessment and life-cycle cost
savings calculation by Booysen et al. (2021) provides literature background information for the
environmental and energy cost evaluation undertaken in this study.

2.4.1 Occupancy-based light control

According to Roisin, Bodart, Deneyer and D’Herdt (2008), lighting energy savings of
occupancy sensors depend on the occupancy rate. In other words, light energy savings
increase with a decrease in occupational rate. However, the study emphasises that the
relationship between occupation rate and energy saving is not linear due to the sensor
response time delay. The study compared the potential of lighting energy savings in office
rooms using occupancy and daylight-linked control systems, and one of the key findings was
that lighting energy saving is generally higher when occupancy control is supported with
daylight-linked control as well. For such an intervention, savings can vary from 45 to 60%,
depending on building orientation and occupancy rate. In addition, lighting energy savings are
higher when on/off occupancy-based control sensors are used compared to dimming sensors.

In contrast to Roisin et al. (2008), De Bakker, Aarts, Kort and Rosemann (2018) argue that
switching off lights in shared and single offices using an occupancy-based control device
results in non-uniform illuminance distribution and an uncomfortable environment. Their
findings show that using highly granular occupancy-based dimming sensors to control lights
automatically reduces lighting energy in shared office space without impairing visual comfort.
The study also investigated the feasibility of using granular occupancy-based light control on
task, surrounding and background areas at various dimming levels. This was based on an
open-plan office staged in a darkened laboratory room of 5.10 x 5.55 m with male and female
participants between the age of 18 to 65 (de Bakker et al., 2018).

The findings show high energy saving potential for all combinations of dimming levels, but the
minimum dimming level in the background offers the most energy savings and visual comfort
(ibid.). The study discloses that the obtained energy savings are influenced by the research
method, such as the office size, layout and an assumption that absence from the office was
only during lunch hours. Based on the findings of Chang and Hong (2013:29), the absence
assumption is not realistic, mainly because absence from the office during the day varies from
10 to 29 minutes. Chang and Hong (2013:29) also discovered that the daily number of
occupant absences and absence-duration significantly impacts lighting energy consumption
and therefore contributes significantly to discrepancies between measured and simulated
lighting energy saving.

22
Chang and Hong (2013) corroborate the findings of Saha, Florita, Henze and Sarkar (2019)
that occupancy-based control is a crucial element of smart buildings where it adjusts settings
according to occupant preferences as a fundamental human-machine interface due to a high
proportion of time spent indoors. Saha et al. (2019) further outline three major algorithms for
estimating building occupancy. The first one is detection which determines whether an
occupant is present inside a particular space in a building or not. Secondly, the counting
algorithm focuses on the number of people present in a given area or space in the building at
a specific time. Lastly, the tracking algorithms are related to occupants’ movement trajectory
of a particular occupant inside the building (Saha et al., 2019). It is worth noting that these
algorithms equally apply to other energy-use systems of a building, such as HVAC systems.

In the above studies, contradictory views on the most effective application of occupancy-based
lighting were expressed. Based on the views of Roisin et al. (2008) and De Bakker et al.
(2018), automated on/off occupancy-based light control is most effective for single-offices
while dimming occupancy-based is suitable for shared-offices. The effectiveness of both
techniques in this context is related to occupants’ perception and visual comfort, which also
influence the energy saving potential. Another crucial aspect of occupancy-based light control
is the occupant sensing algorithm. As defined by Saha et al. (2019), the sensing algorithm can
also determine the suitable location of the occupancy-based light control. For example,
detection sensing automatic on/off occupancy-based light control will be more effective in a
single office. Similarly, tracking sensing dimming occupancy-based light control will result in
more effective light control in a large shared-office space.

Given that occupancy-based light control is one of the commonly lighting energy efficiency
interventions, the studies appraised provide the fundamental understanding and principles of
applying occupancy-based lighting interventions in a building. The shortfalls of automatic
on/off and dimming occupancy-based light control were distilled in order to serve as a guide
in selecting lighting energy efficiency interventions in this study.

2.4.2 Daylight-linked control

In recent years, daylighting has proven to be an effective and useful strategy in modern
architecture to achieve energy efficiency, visual comfort and user-friendly buildings (Alrubaih,
Zain, Alghoul, Ibrahim, Shameri and Elayeb, 2013). Alrubaih et al. (2013) also note that
daylight is the best light source for good colour rendering as it is closer to the human visual
response. The study also emphasises that daylight in a building alone does not result in energy
savings (ibid.:494). Integration of daylight-linked control with electric lighting is required to
achieve energy efficiency through optimisation of daylighting (Alrubaih et al., 2013). Based on

23
this concept, Lim, Keumala and Ghafar (2017) explored the energy saving potential and user-
behaviour of daylight-assisted task lights in a green design office building and overhead
electric lights in a conventional design office in Malaysia. Their study argues that energy
security and environmental sustainability can be achieved using task lights in office space. A
quantitative approach that measures the ambient daylight level and indoor electric light level,
as well as a qualitative method focusing on occupants’ perceptions, were adopted in the
research. The study quantifies lighting energy consumption using lighting power density
(LPD), which was defined as the total wattage of lights installed per unit floor area. The study
finds that task lights alone reduce the LPD by 0.60 W/m2 while integrating daylight-linked
control results in 3.04 W/m2 reduction. Based on this finding, the study recommends that, for
optimum energy savings, task lights could be combined with daylight-linked control systems.
In addition, the study indicates that occupants expressed satisfaction with using task lights for
paperwork and colour rendering but not for computer-related tasks.

The studies appraised in this section demonstrate the significance of daylighting towards
energy efficiency. However, inappropriate integration of daylighting in a building can
negatively impact building energy performance. Chi, Moreno and Navarro (2018) stress that
in addition to visible radiation, sunlight consists of infrared and ultraviolet radiation that may
result in overheating and glare when users are exposed to excessive sunlight. The study
argues that the allowable amount of daylight within a space should be balanced to achieve
adequate energy savings. Given that there are no benchmarks for daylight availability area,
on-site energy used for artificial light supply, heating and cooling, the study also investigated
the available daylight on the work plane for overall building lighting, heating and cooling
energy. The study was therefore able to demonstrate a linear relationship between cooling
energy consumption and the over-lit area. (Chi et al., 2018) and therefore recommended that
confining over-lit areas to less than 40% in the south zone of a building and less than 50% in
the north, east and west zones could limit overall energy use intensity to less than 120
kWh/m2/year within the warm and temperate climatic region of Seville, Spain.

The technical feasibility of utilising daylight for lighting energy efficiency was clearly
demonstrated in both studies reviewed in this section. However, the approach of using daylight
for electrical energy saving and the possible consequence of excessive daylighting was the
focal point of this appraisal. Daylight-linked control would therefore be required to achieve
lighting energy savings due to daylight. In this sense, the electric lighting energy consumption
is determined by the available daylight level. The review of additional studies on the
contribution of daylight toward energy savings is covered in the next section. Daylighting

24
design poses the risk of overheating during the summer season and glare in the winter season
due to the low-angle sun rays. Both effects can lead to excessive energy usage for space
cooling and lighting, thereby defeating the overarching goal of design optimisation for
daylighting.

As a background study for this research, Lim et al.’s (2017) research provides the daylighting
energy saving principle that was adopted in this research, while Chi et al. (2018) provided an
understanding of the consequence of over-lit space, which thus guided the analysis for areas
of the case study building with excessive solar exposure.

2.5 Estimating Energy Savings of Lighting Interventions

Most building owners or businesses are motivated to adopt effective daylighting interventions
once proven to reduce energy consumption and related utility costs (Krarti et al., 2005 and
Lee et al., 2018). Estimation of energy savings of lighting interventions is therefore crucial in
promoting lighting energy efficiency through daylighting optimisation or occupancy monitoring.
Yu et al. (2015) demonstrate the estimation of potential energy saving from daylight using field
measurement, software simulation, empirical formulae and estimation algorithms. Field
measurement involves the use of photometric sensors and energy meters to quantify the
energy savings of daylighting in a building. The study indicates that due to the expense and
duration required for field measurement, extrapolated data for a one-year cycle of local
weather conditions can be adopted.

In a previous study, Atif and Galasiu (2003) adopt field measurements to investigate daylight
impact and electric lighting usage of two atrium buildings. An image of one of the buildings is
presented in Figure 2.6 (a). Their study questioned the common hypothesis that ground floors
in atrium-lit buildings receive sufficient daylight to effectively displace electric lighting. Two
daylight-linked control systems, continuous dimming and automatic on/off were used to
monitor the daylighting energy performance of two case study buildings in Canada. For
selected months in winter and summer, indoor horizontal illuminance was monitored at
strategic indoor spaces in both buildings. The approach yielded data around 25 test points
daily from 7h00 to 19h00 under automatic on/off lighting control and 39 test points under
continuous dimming lighting control. Night-time lighting illuminance was also monitored and
used to obtain the actual daylight contribution by subtracting day-time illuminance from night-
time illuminance. In both buildings, electric light burning hours and power profiles were
monitored as well in winter and summer seasons.

25
The study found that horizontal illuminance at ground floor perimeter areas was above 200
lux between 9h00 and 13h30 in winter and mostly above 500 lux between 9h00 and 17h00 in
summer. In both seasons, the illuminance of areas that are 3.5 m away from the perimeter
was below 200 lux during the same period. The study concluded that continuous dimming light
control was the most viable lighting energy savings technique. During the occupancy period
(08h00 to 18h00), the study found that continuous dimming daylight-linked control resulted in
60% lighting energy savings in winter and 74% in summer. The study attributes the poor
performance of automatic on/off lighting control to the oversized lighting system of the building
and built-up snow on the glazed roof. Based on the findings, the study further argues that
daylight distribution in buildings and the type of automated lighting control are crucial for
daylight-linked energy savings.

Figure 2.6: Daylight-linked automatic lighting control: (a) atrium building in Quebec City,
Canada (Atif and Galasiu, 2003:443) and (b) passive solar design office building in Kuala
Lumpur, Malaysia (Lim et al., 2017:169)

As discussed in the previous section, Lim et al. (2017) adopted field measurement to explore
the energy saving potential and occupants’ behaviour towards daylight-assisted task lights in
two office buildings in Malaysia. Due to limited monitoring equipment, the study monitored the
office buildings for two consecutive months (September and October 2015). The approach
was motivated by the fact that sky conditions in Malaysia are usually intermediate for 85.6%
and overcast for 14.0% of the time. Based on local weather and long-term climatic data, there
were no significant variations in overcast and rainy days for both months. The study used
Hobo-U12 loggers to monitor daylight and electric lights’ illuminance while the electric light
energy consumption was derived through LPD (installed lights wattage divided by floor area)
(ibid.).

26
Software-based simulation is commonly applied in studies evaluating building daylighting and
potential energy savings. However, Yu et al. (2015) argue that, despite the shorter period in
software simulation, its accuracy is a major concern and should be validated through
measurements in order to ensure reliable outcomes. The study adds that simulation is
instrumental in the early design phase when daylighting features of buildings can be
manipulated for optimised outcomes. Some of the commonly used software applications are
EnergyPlus (Lee et al., 2018), DaySim (Akimov, De Michele, Filippi Oberegger, Badenko and
Mainini, 2021), DOE-2.1 (Asif Ul Haq, Islam, Shihavuddin, Maruf, Al Mansur and Hassan,
2020), IES-VE (Freewan and Al Dalala, 2020) and Autodesk Revit (Kota, Haberl, Clayton and
Yan, 2014).

Several studies are also based on empirical formulae derived from prior/historical studies and
commonly applied in building lighting standards (Yu et al., 2015). For example, the European
Standards En15193 (European Committee for Standardisation, 2017) provides a calculation
method for indoor lighting energy requirements relative to daylight. Lastly, estimation
algorithms are manual calculation procedures computed by researchers for fast and accurate
daylight potential energy saving (Yu et al., 2015). Some of the algorithms highlighted by the
study are “calculation using the cumulative frequency distribution of daylight illuminance,
calculation using mathematical integration, calculation algorithm of artificial light output by
considering the contribution from other luminaires and calculation based on the property of
daylight responsive dimming system” (ibid., 2015:500).

Quantifying electric light energy savings due to daylight is quite complex because both
concepts use different energy sources. In addition, effective daylight in a building is influenced
by several factors such as tasks, building design, ambient sun ray blockage, local weather
conditions, human factors, and floor arrangement. As highlighted by Yu et al. (2015), field
measurements, software-based simulation, empirical formulae, calculation, and estimation
algorithm are some of the basic approaches used for estimating lighting energy savings due
to daylighting. The experimental aspect of this research is based on field measurement, while
a software-based approach was adopted to evaluate and quantify the whole-building
daylighting distribution and energy performance.

2.6 Conclusions

Based on the studies appraised in this Chapter, daylighting design strategies, weather and
surrounding obstacles were identified as the main influencing factors of daylighting in
buildings. Side-lighting and top-lighting emerged as the two fundamental daylighting design

27
strategies in buildings. Although top-lighting was deemed the best daylighting strategy, it is
not encouraged in climates with a high cooling load. One of the key insights on lighting energy
efficiency is that replacing inefficient light fixtures such as T8 fluorescent tubes with T5
fluorescent tubes or LED constitutes a substantial intervention. However, such measures pose
a performance and investment risk for ESCos in performance contracting projects. It was also
clear that integrating energy efficient lights with occupancy-based or daylight-linked control
systems increases lighting energy savings. However, occupancy-based sensor operating
mechanisms (on/off and dimming), response time, and occupancy rate significantly influence
the amount of energy reduction. Dimming of task, surrounding, and background area lights
were revealed to have the most energy savings performance while also ensuring visual
comfort of occupants compared to on/off control devices.

In addition, the reviewed studies highlight that the integration of daylight in a building does not
automatically constitute energy saving without daylight-linked control. The negative impact of
daylighting, such as glare and overheating due to inappropriate design, was also highlighted
in several studies. In quantifying daylighting and potential energy savings performance, some
of the methods recommended in some of the studies include field measurement, software-
based simulation, empirical formulae and estimation algorithms.

28
 3CHAPTER 3

METHOD OF THE STUDY - BUILDING DAYLIGHT LEVEL MEASUREMENT, ENERGY

PERFORMANCE ESTIMATION AND SIMULATION

3.1 Introduction

A quantitative research approach was adopted for this study. According to Kruisselbrink,
Dangol and Rosemann (2018), measurement of daylight levels in a building can be based on
spot or continuous methods. In the spot method, a measurement grid covering the overall
lighting space is established. Gentile, Dubois, Osterhaus, Stoffer, S., Amorim, C.N.D., Geisler-
Moroder, D. and Jakobiak (2016) suggest that spot measurement of daylight level should only
focus on relevant task areas for simplicity and reduced cost.

Unlike spot measurement, continuous methods span over a long period and involve dividing
the space into daylit zones (Kruisselbrink, Dangol and Rosemann, 2018) or area of interest.
The area of interest is determined by the proximity from the perimeter walls but is usually set
at 0.5 m from the perimeter wall. The study recommends that at least one measurement point
should be located at the working plane level for continuous daylight measurement in an office
space. In addition to measurements, Yu et al. (2015) highlights software simulation, empirical
formulae and estimation algorithms as key approaches used to estimate the energy savings
potential due to daylight. This study adopted a continuous method for the experimental
daylight level measurement for a heritage category A3 building. Field measurement and
software-based simulation were also used to compute the potential for optimised energy
savings due to the difference in energy consumption levels between “as-is” versus
performance after daylighting interventions.

This Chapter therefore presents the procedure and instrumentation adopted in the continuous
field measurements and software-based simulation of the building daylighting and energy
performance. To this effect, the Chapter also presents the building description as well as the
experimental and simulation-based approaches used. This is followed by a discussion of
electric lights and appliances in the building and mathematical formulae used in quantifying
the lighting energy performance of the building.

Figure 3.1 presents an overview of the overall research process, while Table 3-1 presents a
summary of primary data needed, collection tools, presentation, and analyses, as well as
sources of secondary data used in the study.

29
Figure 3.1: Workflow of the research process

30
Table 3-1: Summary of data needed, collection tools, and analyses relative to the main research question
Research question Data collection tools and process Data analysis approach
Sub-question 1: What is the status quo EUI for Primary data: The floor area of selected offices was measured to The light energy consumption was computed by multiplying the number of lamps
the selected case study building? compute the “as-is” lighting energy consumption of the case study in the selected offices by their wattage and annual building hours. The building
building. IES-VE application was used to simulate the entire model, in-built weather data of the building location and schedule operation
building “as-is” EUI, and the dimensions of the building walls, hours, lighting and appliances were used in the IES-VE application to simulate
windows, doors, and floor area were used to develop the building the “as-is” EUI of the entire building. The selected offices’ energy consumption
model. and whole-building EUI were presented and analysed using bar and pie charts.
Secondary data: The power ratings on the nameplate of the Tables were also used to summarise the findings. The findings were compared
installed lamps in the offices and their annual burning hours were with published academic findings for justification. The graphic representation of
also used. The various offices’ occupant class schedule was used the data and analysis were presented and analysed in Chapter 4.
to determine the burning hours of the selected office and used to
extrapolate that of the entire building.
Sub-question 2: What are the key daylighting Secondary data: Non-invasive lighting interventions were The lighting interventions were implemented using the IES-VE, guided by the
interventions of the case study building? considered in this research. The implemented interventions are typical daylight-linked control principle from literature studies.
daylight-linked control, occupancy control, and a combination of The impact of the lighting interventions were demonstrated with the daylight level
both controls. and EUI of the building. Both parameters were presented in line graphs and bar
charts. A comparative analysis between “as-is” and with intervention EUI was
also tabulated. Sub-findings and graphs of the key lighting interventions were
covered in Chapters 4 and 5.
Sub-question 3: What is the “as-is” contribution Primary data: Onset HOBO U12-012 photometric sensor was The measured indoor daylight level of the offices, as well as the outdoor ambient
of daylighting to the EUI of the building? used to monitor the indoor daylight level in selected offices in the daylight horizontal illuminance and solar global horizontal irradiance, were
building. The influence of the outdoor ambient weather conditions presented and analysed using line graphs mainly. Simulated daily movement of
was also measured. Likewise, the installed lamps in the offices the sun, solar exposure, luminance and illuminance distributions, and photos of
were counted to deduce the building’s lighting energy savings due the building was also used to depict the daylighting of the building. Lastly, bar,
to daylighting. pie charts and tables were used to analyse the computed selected offices’ lights
Secondary data: Daylight-linked lighting control principle was energy consumption and the simulated EUI of the entire building. Analysis and
assumed from literature studies to operate the indoor electric sub-findings of daylighting contribution to the building EUI were presented and
lights in the offices with respect to measured indoor daylight level.discussed in Chapter 4.
Sub-question 4: What is the potential lighting Primary data: The IEV-ES application was used to implement the The operational cost and environmental impact were deduced using the current
retrofit interventions towards reducing EUI, and lighting retrofit interventions. Thus, the dimensions of the building, energy tariff of the university and local coal-fired power plant emission factors.
what would be the viability of such as explained above, were used to develop the building model. Simulated daylight profiles of the building were presented in line graphs.
interventions? Secondary data: The daylight-linked and occupancy lighting Average daylight luminance, illuminance, daylight autonomy, and useful daylight
control operating principles were obtained from literature studies. illuminance were used to analyse the light retrofit interventions of the building.
The IES-VE application building materials thermal-optical Also, simulated EUI of the building was presented in bar and pie charts. The “as-
properties database was used to assign optical properties to the is” and with-intervention EUI were compared using bar charts and tables. The
building. analysis and discussion of the results were covered in Chapters 4 and 5.
Main research question: What is the contribution of daylighting towards EUI-reduction in
The main research question was answered by answering all the sub-questions. The sub-findings and
heritage Place-of-Instruction buildings in South Africa, and what are the possible lighting
analysis, as well as the discussion of results, were presented in Chapters 4 and 5. Overall findings in
interventions towards reducing EUI and enhanced financial and environmental benefits of
response to the research question are consolidated in Chapter 6.
such buildings under energy zone 4 as per the 2nd edition of SANS10400-XA criteria?

31
3.2 Description of the Case Study Building

Livingstone Hall, located at the University of Fort Hare in Alice, is used as a case study in this
research. Although several departments are based in the building for their respective
academic activities (Geology, Physics, Computer Science and Mathematics), the study
focuses mainly on the Physics Department section of the building. In reference to EPC
regulations, the building falls under category A3 (place-of-instruction) (South African Bureau
of Standards, 2021). In addition, the building was built more than 90 years ago (completed
mid-1930s) and is thus considered a heritage building under the Heritage Resources Act
(South Africa Heritage Resources Agency, 1999). Photos of the Physics Department are
presented in Figure 3.2.

Figure 3.2: The Physics Building at the University of Fort Hare; (a) front view and (b) rear
view

The Physics Department occupies a two-story portion of the building, with the front view shown
in Figure 3.2 (a) and the rear view in Figure 3.2 (b). During the period of this research, the
building consisted mainly of laboratories, lecture halls, single offices and shared offices on
both floors. Technical details of the building, such as the floor plans, facades and solar
exposure analysis, are presented graphically and discussed in Sections 3.3.2 and 5.2.1.

32
3.2.1 Data needed, collection and analysis approaches
As highlighted in the previous section of this Chapter, for the experimental approach,
photometric sensors were installed in selected offices in the building to monitor the indoor light
levels, while Integrated Environmental Solution-Virtual Environment (IEV-VE) software was
used for the simulation approach. For simulation inputs, the building dimensions and envelope
properties were used to develop a model of the building as required for simulation. Even
though the use of both approaches simultaneously was not intended for comparison of
outcomes, the simulation-based approach facilitated a holistic daylighting analysis as well as
whole-building energy performance evaluation, while on the other hand, experimental
measurement was used to obtain real-time daylight levels of a selected sample of offices in
the building due to limited sensors.

3.2.2 Overview of the Installed Instrumentation

The University of Fort Hare is located in the town of Alice, Eastern Cape, at latitude 32°47’7”
S and longitude 26°50’43” E, and at 540m above sea level. Alice is located within energy zone
4 of South Africa as per SANS10400-XA (South African Bureau of Standards, 2021). GIS
imagery of the University of Fort Hare, Livingstone Hall and a photo of the Physics Department
(building) are presented in Figure 3.3

Figure 3.3: GIS imagery of the University of Fort Hare, Alice in Eastern Cape (a); imagery of
Livingstone Hare at the University of Fort Hare, and a photo of the Physics Building

In the experimental aspect of the research, the indoor daylight level, ambient daylight, and
solar global horizontal irradiance were measured using Onset HOBO U12-012 photometric

33
sensor, LI-Cor LI-210R photometric sensor, and Kipp & Zonen CMP 11 pyranometer,
respectively. The HOBO U12-012 and a solar weather station set-up are shown in Figures 3.4
and 3.5, respectively.

Figure 3.4: Onset HOBO U12-012 Temp/RH/Light/Ex sensor

Figure 3.5: Solar radiation and weather station at the SolarWatt Park, University of Fort Hare

In the absence of dedicated submetering for the building or lighting energy use, the lamps in
each office and other parts of the building were physically counted in order to estimate the
“as-is” lighting energy consumption of the building. The lamps' wattage and annual burning
hours were also taken into consideration. Lastly, the dimensions of the building walls,

34
windows, doors, and floor area used for modelling the building were measured using a
measuring wheel and tape.

Secondary data used in the research include the lighting requirements for typical office tasks
in South Africa as stipulated in the Department of Labour (1993). Daylight-linked lighting
control operating principles used to simulate the operation of the electric lights in the offices
were distilled from existing studies. The inscribed power rating of the lamps was used together
with the number of lamps and annual burning hours to determine the lighting energy
consumption of selected offices in the building. In terms of the whole-building EUI simulation,
the class timetabling, and typical staff schedules were used to establish the annual burning
hours based on typical class schedules. The inscribed power rating and the burning hours
were therefore part of the secondary data used in the research.

GHG emissions factor for coal-fired power generation in South Africa and the energy tariff
structure used to deduce the environmental impact and operational cost, respectively, were
obtained from Eskom (2020) and Eskom (2021). The selection of optic properties of building
envelope components from the in-built IES-VE database was guided by insights from the
studies reviewed. The web-based weather database in the IES-VE application was used to
simulate the building’s daylighting and energy performance. This information from the IES-VE
database, as used in building modelling and simulation, also form part of the secondary data
for the study.

3.3 As-is Daylighting of the Case Study Building

3.3.1 Experimental approach

Continuous measurement (as substantiated in Kruisselbrink et al., 2018) was adopted to
evaluate the “as-is” daylight levels of the building. However, only a purposely selected sample
of single offices was covered in this experiment due to the limited number of photometric
sensors. The offices considered were strategically selected to ensure that the daylighting
performance was sufficiently sampled and represented in terms of the daily trajectory of the
sun for both summer and winter seasons. As shown in Figures 3.6 and 3.7, three office spaces
on each floor (two offices on the west and three offices on the east) were selected for the
study.

35
 Figure 3.6: The ground floor layout of the Physics building indicating selected offices.

Figure 3.7: The top floor layout of the Physics building indicating selected offices

The offices have different floor arrangements with varying distances of desks from the widows.
The average distance of the desk from the windows in the selected offices is 4.50 m, with the
longest distance at 6.50 m in space 22. Figures 3.6 and 3.7 also show that most of the offices
use vertical fabric (blue) blinds, while venetian aluminium blinds were installed in space 41.
For purposes of the “as-is” study, occupants were not constrained on how to operate the blinds

36
or lights. Instead, daylighting levels were evaluated under actual operating conditions by the
users/occupants. In line with Kruisselbrink et al. (2018), daylight levels at offices’ desks as the
work plane were targeted in the experiment. The Onset HOBO U12-012 Temp/RH/Light/Ex
sensor was placed on the desk of the selected offices. The HOBO U12-12 sensor is designed
for indoor light measurement, ranging from 1 lux to 48 438 lux (Onset, 2018). The sensor is
powered by a 3V button battery and has in-built memory capable of storing up to 43 000
measurements (see Figure 3.4 for a photo of the HOBO U12 sensor used in this experiment).
The measurement was conducted from April to July, where April and May were considered as
(late) summer season, while June and July represent the winter season. The data were logged
in 30 minutes intervals and downloaded every fourth week.

In addition to the selected offices, light level measurement, outdoor ambient solar radiation
and horizontal illuminance were measured. The outdoor ambient solar radiation was used to
establish clear and cloudy sky conditions to allow for indoor daylight analyses relative to
available outdoor ambient horizontal illuminance under clear and cloudy sky conditions. Data
from the solar weather station at the SolarWatt Park (see Figure 3.8) at the University of Fort
Hare were also used to achieve this objective.

Figure 3.8: Solar weather station at the SolarWatt Park, University of Fort Hare

The SolarWatt Park is less than 500 m from the case study building. The weather station
entails a SOLY Gear Drive dual-axis sun tracker, which houses a pyrheliometer, shaded
pyrgeometer, pyranometer, and a shaded pyranometer. The pyrheliometer is used to measure
direct normal irradiance (DNI), the shaded pyrgeometer measures downward longwave
radiation, global horizontal irradiance (GHI) and direct horizontal irradiance (DHI) are
measured with the pyranometer and the shade pyranometer, respectively. In addition, an LI-

37
Cor LI-210R photometric sensor was also installed at the weather station to monitor outdoor
horizontal illuminance. The specifications of the radiometers are summarised in Table 3-2.

Table 3-2: Radiometers specifications (Overen, Meyer and Makaka, 2021)

Radiometer Sensitivity Spectral range Response time
Pyranometer 5 – 10 µV/W/m2 0.3 – 2.8 µm <5s
Li-Cor LI-210R 30 µA/100 klux <1 µs

The specifications of the pyranometer and Li-Cor LI-210R used for GHI and HDI measurement
are necessary to show the reliability and measurement range of the sensors. The
specifications are also used to validate the reported data and findings.

3.3.2 Simulation approach

One of the shortfalls of the experimental approach is the limited number of spaces covered
due to insufficient sensors. A simulation approach was therefore adopted to analyse the
whole-building daylight distribution. A walk-through audit of the building was conducted to
obtain the dimensions of the building envelope components, which served as inputs for
developing the building model. Autodesk Revit 2021 was used to capture the model, which
was then exported to IES-VE, where it was fully developed as the key input for simulation. The
Autodesk Revit drawings of the floor layout of the building are presented in Figures 3.9 and
3.10.

38
Figure 3.9: The ground floor layout of the Physics building at the University of Fort Hare

39
 Figure 3.10: The top floor layout of the Physics building at the University of Fort

The building is oriented 164°S and has a total floor area of 1 170.44 m2. This includes the
Teaching and Learning Centre, which is outside the scope of this research. Moreover, the floor

40
area of the various rooms on both floors varied. The average height of the rooms on both
floors was 3.00 m. Figure 3.11 shows the various elevations of the building.

Figure 3.11: The various elevations of the building

According to Bradshaw (2006:279), the size, position and orientation of windows are important
factors that influence the daylight levels in a building. The sizes and positions of the windows
in the Physics Department building were therefore noted as part of the data for the study. As
shown in Figure 3.12, all the windows are on the various facades (vertical windows) but no

41
roof windows (skylight, clerestory or monitor windows). The windows were also noted to be of
two main sizes, as shown in Figures 3.12 (a) and (b).

Figure 3.12: The two main window sizes in the building; (a) small and (b) large

Windows in both sizes were made of steel frames with a white-paint finish on inner and outer
surfaces. The large windows have an average glazing area of 1.00 m2 and are mostly used
on the south elevation where the laboratories and lecture halls are located. The east and west
elevations, where most of the offices are situated, mainly use small windows with an average
glazing area of 0.80 m2. In addition to the large and small windows, another set of windows
with an average glazing area of 1.5 m2 was installed on the top floor of the north elevation, as
shown in Figure 3.11.

Glazing for all the windows is single pane 6 mm clear glass. The assumed properties of the
windows are presented in Table 3-3. The glass was also assumed to be isotropic in behaviour,
i.e., the properties shown in Table 3-3 apply to both sides of the glass.

Table 3-3: Thermophysical properties of glass as adapted from IES-VE building material
properties database
Glass type Soda-lime
Thickness 6 mm
U-value 3.76 W/m2K
Emissivity 0.84
Young’s modulus 70.0 ×103 MPa
Specific heat 1.0 ×103
Reflectance 0.072
Reflective index 1.53

42
The IES-VE 3D SunCast model of the building is presented in Figure 3.13.

Figure 3.13: A 3D SunCast model of the case study building on the 15 July

As shown in Figure 3.13, the building model includes adjacent buildings (pink structures) and
trees that could be potential sun blockades due to their proximity and height. Bisho weather
data was used in the simulation since the Alice weather data is not in the IES-VE weather
database. However, Bisho is approximately 70 km away from Alice, and both towns are
categorised under energy zone 4 (South African Bureau of Standards, 2021). Further,
parameters such as solar exposure, luminance, illuminance, daylight factor, useful daylight
illuminance and special daylight anatomy were employed to analyse the whole-building
daylighting performance in Chapter 4 of this report.

3.4 Lighting Energy Performance of the Building

3.4.1 Electric lights and appliances in the building

The walk-through audit of the building also covered electric lights and appliances. In the audit,
the number and nameplate power ratings of the lights and appliances in the respective spaces
of the building were obtained. The data collected were used to calculate the light energy
consumption of the selected offices and also as input into the building model to simulate the
EUI of the entire building. During the audit, electronic and magnetic fluorescent fixtures were
noted to be the most common. The fixtures used were Hyundai YZ-T5-3-band 35W 6400 k
tubes, while Osram Eco ST8E-1.5m 20W tubes were used for the magnetic fixtures. Photos
of the Figure 3.14 shows the types of fluorescent tubes used in both fixtures.

43
 Figure 3.14: Electronic and magnetic fluorescent tubes used in the building.

A total of 332 fluorescent tubes were recorded in the audit. The inventories for ground and top
floor lights and appliances are presented in Appendix A-1 and A-2, where an average of one
desktop or laptop personal computer was noted for the single offices. A projector was installed
in each of the lecture halls. Several electrical equipment and devices were also noted in the
laboratories. However, all the laboratory electrical equipment was grouped as lab apparatus
for simplicity in modelling and simulation. It was also noted that the building has no central
heating, ventilation or air conditioning system, although there were standing fans, heaters and
split-unit air conditioners in some offices.

3.4.2 Occupancy and operating hours

The energy performance of the building was achieved based on calculation and simulation
due to the lack of dedicated-metering energy data. A crucial aspect of calculating and
simulating the energy performance was to establish the building occupancy period, lights and
time-of-use of appliances. The normal occupancy schedule of the building before the
Coronavirus (Covid-19) outbreak was assumed based on the Faculty of Science and
Agriculture 2018 prospectus that contains the class schedules for the building. The class
schedule was then used to estimate the lecturers’ and students’ occupancy periods, as well
as the operating hours for lights and appliances.

3.4.3 Estimation of lighting energy for Selected offices

The field measurement approach (Yu et al., 2015) was adopted in estimating the lighting
energy savings due to daylight in the selected offices. In this approach, the electric lights in a
given space are controlled with a daylight-linked sensor. The energy saved while the electric
lights are switched off by the sensor as a result of sufficient daylighting in the space is,
therefore, the energy savings due to daylight (Lim et al., 2017). The same principle was
adopted in the experimental evaluation of the light energy savings due to daylighting in the
selected offices, even though daylight-linked sensors were not installed. Equation 1 was
therefore used to determine the electric light energy consumption.

44
 n
ε lit
= ∑ (χ µ l1 f + χ l2 µ f ... + χ ln µ f )
l =1,2,3..., n (1)

Where ε lit is the total electric light energy in a given space (kWh), χ l is the nameplate power

(W) rating of the lights, µ f is the time in hours the lights are switched on, and n is the total

number of lights in a given space. From Equation (1), the operational cost of the selected
offices due to electric lights was derived using Equation (2).

αt
χ=
R ε lit ×
100 (2)

where χ R (R) is the daily operational cost of the selected offices due to electric lights usage,

while α t (c/kWh) is the energy tariff. The energy tariff is given by the national utility and varies

with consumers, transmission, time of use, and season.

The light energy savings due to daylight in the selected office was therefore derived by
assuming the effect of a daylight-linked sensor. That is, the sensor switches off the lights when
the indoor illuminance is equal to or greater than 300 lux and on during occupied hours when
the illuminance is below 300 lux (the light level benchmark for conference rooms and general
office tasks in South Africa). In this case, the daylight-linked controlled electric light energy

consumption can be denoted as β lit (kWh) and derived as shown in Equation (3).

n
βlit
= ∑ (χ µ l1 f + χ l2 µ f ... + χ ln µ f )
l =1,2,3..., n (3)

The average daily electric lights energy savings due to daylighting is given as shown in
Equation (4);

φsav
= ε lit − β lit (4)

The higher the value of φsav , the better the daylighting performance of the offices. If φsav equal

to zero, it implies that the daylight level in the offices cannot be utilised for typical office tasks.

By replacing ε lit with φsav in Equation (2), the financial benefit due to daylighting was obtained.

45
3.4.4 Whole-building energy use intensity simulation
The energy-use profiles based on the use-periods as discussed in Section 3.4.2, as well as
the lights and appliances wattage in Appendix A-1 and A-2, were entered into the building
model in the IES-VE application for simulation. The simulation was set to analyse the whole-
building energy for a full year while recording the data in 30 minutes intervals. Once
completed, a second energy profile with the daylights’ interventions was loaded into the model
with a different title but simulated with the same settings. The “as-is” EUI of the building was
obtained from the initial simulation, while the optimised EUI was generated from the second
simulation.

3.5 Data Collection, Presentation, Analysis, and Discussion

Data collection in this research was guided by the sub-questions in order to achieve the
objectives of the study (refer to Chapter 1, Section 1.6). The above sections highlighted the
key activities and procedures used in collecting primary and secondary data. Secondary data
were sourced from journal articles by Yu et al. (2015) and Roisin et al. (2008), which provide
detailed approaches for implementing simulation-based lighting energy efficiency intervention
in a building. The articles therefore served as secondary data sources towards addressing
sub-question 4. Similarly, the IES-VE in-built building materials properties database was
adopted as a source of secondary data for simulating the “as-is” and optimised whole-building
daylight distribution and EUI as needed in sub-questions 1, 3, and 4. The walk-through audit
presented in Section 3.4.1 was another vital source of secondary data used in addressing sub-
question 1, 3, and 4 as well. Other sources of secondary data are the University of Fort Hare,
Faculty of Science and Agriculture 2018 prospectus, and the Eskom tariffs & charges booklet
2021/22 (Eskom, 2021).

In terms of data presentation, collected data pertaining to all sub-questions were visualised
using line graphs, pie charts, bar charts, and 3D models of the building and presented in
Chapters 4 and 5 of this report. Tables summarising analysed or computed data were also
adopted. The daylighting performance of the building was analysed using parameters such as
luminance, illuminance, daylight factor, useful daylight illuminance and spatial daylight
autonomy. Average space wattage and EUI were used to analyse the energy performance of
the building. The operational cost of the building relative to energy usage and environmental
impact were analysed with energy cost in Rand and greenhouse gas emissions such as CO2.

Data presentation and analysis for sub-questions 1, 2, and 3 were covered in Chapter 4, while
Chapter 5 covers data analysis for sub-question 4. In both Chapters, the analysed data were

46
discussed in terms of the reasons and physical implications of the data pattern. The discussion
also draws a comparison of the obtained data with other similar published findings. A summary
of the data collection, presentation, analyses and discussion was presented in Table 3-1.

3.6 Ethics Consideration

All the data collection processes and tasks were conducted in line with the guidelines on ethics
in research by the University of the Witwatersrand. Permission from the relevant authorities of
the University of Fort Hare was obtained to install data acquisition systems in the building and
to gain access to the building for data collection periodically. In addition, the occupants of the
offices were duly informed and gave their consent before the photometric sensors were
deployed as well as any other measurements. As agreed with the occupants, none of their
personal information or activities was disclosed in this report or in any future publications.
Lastly, in line with the ethics waiver obtained for the research, no human participation in the
form of a questionnaire or interview was involved in the research

47
 4CHAPTER 4

AS-IS DAYLIGHTING AND POTENTIAL ENERGY SAVINGS OF THE BUILDING

4.1 Introduction

Furthermore, several metrics have been adopted over the years to assess daylighting in a
building. According to Yu et al. (2015:495), daylight illuminance (lux) is the most commonly
used metric for evaluating the daylight performance of a building. Daylight illuminance is often
used together with luminance (cd/m2) (Szokolay, 2008). Illuminance is measured horizontally
at a given work plane and daylight zone or area of interest (AOI) to determine workplace light
level requirements (Kunwar, Cetin, Passe, Zhou and Li, 2020). The work plane is equivalent
to the height of an office desk, which ranges from 0.70 to 0.85 m, while AOI is usually 0.5 m
from the boundary walls and windows (Kruisselbrink, Dangol and Rosemann, 2018).

This Chapter presents an evaluation of the “as-is” daylight performance of the case study
building based on the daylighting factors reviewed in Chapters 1 and 2 and the method
presented in Chapter 3. Lighting energy consumption and potential energy savings due to the
building’s “as-is” daylight performance were also analysed. The Chapter starts with the
experimental-based daylight illuminance analyses of selected offices in the building and their
respective lighting energy performance. Thereafter, the simulated whole-building daylight and
energy performance analyses are presented, followed by the derivation of sub-findings to sub-
questions 1, what is the status quo (as-is) EUI for the selected case study building? And sub-
questions 3, what is the “as-is” contribution of daylighting to the EUI of the building?

4.2 Experimental-based Daylighting Performance Analysis of the Building

The primary objective of the experiment was to derive the “as-is” work plane daylight
illuminance of the selected offices (spaces 49, 41 and 43 on the building’s ground floor and
spaces 11, 15, and 22 on the top floor). The daylight illuminance of the spaces was monitored
concurrently with the outdoor ambient solar radiation and horizontal daylight illuminance, as
analysed in the next section.

4.2.1 Outdoor ambient solar radiation and light level

For the purposes of this study, global horizontal irradiance (GHI) was prioritised because it
consists of both direct normal irradiance and diffuse horizontal irradiance. It is therefore
sufficient to illustrate solar radiation availability of outdoor horizontal daylight illuminance (HDI)
versus the resultant indoor daylight performance. The measured GHI and HDI are presented
in Figure 4.1.
48
Figure 4.1: Measured global horizontal irradiance and outdoor horizontal daylight illuminance
in the monitoring period

From Figure 4.1, April and May’s data represent summer GHI and HDI, while winter GHI and
HDI were obtained from June and July. In the entire monitoring period, a total of 3 974 data
entries were captured for GHI and HDI. However, 184 data entries, approximate 5% of the
total data entries, were lost due to the data acquisition system malfunction. The lost data only
occurred in the GHI and HDI, while the indoor illuminance data entries were intact. The missing
data were excluded in all analyses.

The observed average summer GHI was 353.0 W/m2, with a peak of 994.0 W/m2 at 12h00 in
April. The resultant HDI was found to be 36 407.24 lux on average, peaking at the same time
as GHI at 99 250 lux. The obtained maximum HDI is 750 lux less than a typical clear sky
sunlight illuminance, according to Al-Obaidi, Ismail and Abdul Rahman (2015:2). Further,
slightly less irradiance and illuminance were observed in the winter period. The average winter
GHI and HDI were found to be 331.58 W/m2 and 33 457.55 lux, respectively. The maximum
values recorded for both parameters were 811.0 W/m2 and 82 000.0 lux at 12h30 in July.

Although the obtained outdoor HDI in the winter season was lesser than a clear sky sunlight
illuminance by 18 000 lux, Figure 4.1 indicates adequate solar irradiance and outdoor daylight
for effective daylighting in the building, considering that typical workplace light requirement
ranges from 100.0 lux to 3 000.0 lux (Szokolay, 2008:192). However, building designs, human
factors, and adjacent structures are other influencing factors that must be considered.

Further, the sky conditions used to analyse the indoor daylight performance of the selected
offices were deduced from Figure 4.1. This was achieved based on the daily peak GHI and

49
distribution pattern. It was assumed that days with peak GHI greater than or equal to
500.0 W/m2 and uniform GHI distribution were clear sky days, while days with peak GHI lesser
than 500.0 W/m2 with irregular GHI distribution were cloudy sky days. Based on these
assumptions, 19 April and 15 June were considered summer and winter clear sky days,
respectively, while summer and winter cloudy sky days were respectively represented by 22
April and 23 June.

4.2.2 Indoor daylight illuminance

The indoor daylight illuminance in this section covers the “as-is” daylight performance of
selected offices in the building. In other words, no daylighting interventions were implemented
or considered. Also, the daily activities of the occupants in terms of the operation of lights and
windows were not constrained. The indoor illuminance of the selected offices in the entire
monitoring period is presented in Figure 4.2.

Figure 4.2: Indoor illuminance of selected offices in the case study building

The illuminance in Figure 4.2 is a combination of electric light, daylight, and ambient light.
Ambient lights include streetlights, moonlight, and electric light from the hallways. In the
presence of the sun (periods with GHI greater than 0 W/m2), daylight illuminance was found
to be dominant in some spaces. At night (absence of the sun, periods with 0 W/m2 GHI), the
ambient lights were dominant when the electric lights were switched off. A synopsis of the
lighting performance of the selected spaces is presented in Table 4-1, where an adjusted
summer and winter time of the sun’s absence and presence was used to accommodate both
seasons’ data in the same table. From the measured data, summer season sunrise and sunset
50
were 07h00 and 17h30 daily, while sunrise was 07h30 and sunset was 16h30 in winter
season. Hence, the time of the sun’s absence and presence were adjusted for seasons in
Table 4-1.

Table 4-1: Lighting performance of the selected offices in the case study building
Sun absence (18h00 – 6h30) Sun presence (08h00 – 16h00)
Electric lights Actual
Space Avg. Max. Min. Avg. Max. Peak
on duration daylight
(lux) (lux) (lux) (lux) (lux) time
(days) (lux)
April
41 72.64 145.80 3.90 47 113.04 208.90 40.40
12h30
June
43 30.84 177.40 19.70 6 261.75 12775.60 230.91
11h00
June
49 13.13 82.80 3.90 8 52.65 413.90 39.52
11h00
May
11 11.42 27.60 3.90 0 208.40 11711.30 208.40
10h30
April
15 0.20 3.00 0.00 0 202.72 1503.00 202.72
10h30
April
22 24.88 130.11 11.18 9 80.86 224.70 55.98
10h00

The illuminance in all the spaces besides space 15 was above 0 lux in the absence of the sun.
This is attributed to the illuminance of ambient light sources, as indicated earlier. A photo of
the top-floor hallway of the building indicating a source of ambient lights is presented in Figure
4.3.

Figure 4.3: Hallway of the selected ground floor selected offices

51
As shown in Figure 4.3, the upper area, approximately 0.20 m2, of the doors of most spaces
(offices, lecture halls, and laboratories) in the building were made of glass. The glass area of
the door allows light from the hallways into the spaces at night and during the day since the
lights on the hallways were observed to be switched on constantly. Space 15 had a similar
door design as shown in Figure 4.3, but the glass area was covered with cardboard paper to
measure the electric light illuminance without the interference of the lights in the hallway, which
was found to be 185.0 lux. However, the measured electric lights’ illuminance of space 15
cannot be adopted for the other spaces since illuminance is a factor of floor area, the amount
of light in a given space, and the position of the sensors with respect to the external windows.
This is evident in Table 4-1, where the average and maximum illuminances in the absence of
the sun are shown to vary. The actual daylight illuminance of the offices is therefore derived
as the difference between the average illuminance in the presence of the sun and average
illuminance in the absence of the sun, which would primarily be due to electric lights indoors
and in the hallways.

In the presence of the sun, the offices (spaces 11, 15, 41, and 43) on the east side of the
building were found to have the highest daylight illuminance. Except for space 41, the average
daylight illuminance in the east offices was 224.29 lux and 155.89 lux, excluding potential
electric light illuminance. In a similar approach, the average daylight illuminance in space 41
was 40.40 lux, which is five times less than that of the other east offices. Human factors or
behaviour is the major cause of the relatively low daylight illuminance observed in space 41.
A photo of space 41 inner space is presented in Figure 4.4.

As shown in Figures 3.6 and 3.7, space 41 is the only office with Venetian aluminium blinds.
The blinds used in the office prevent a significant amount of sunlight from the inner space, as
shown in Figure 4.4. Despite the poor choice of window cover, opening the blinds during the
day would allow the penetration of sunlight for effective daylighting. As shown in Figure 4.4,
this is not a common practice by the occupants who usually rely on electric lights to illuminate
the office space rather than daylighting allowed by opening blinds. It would be interesting to
establish the reason for such behaviour based on qualitative approach, but this falls outside
the scope of this research.

52
 Figure 4.4: A photo of the inner space of space 41

In contrast, daylight illuminance distribution for spaces 43 and 11 was dominant in both
seasons (see Figure 4.2). The peak illuminance in both spaces was found to exceed 2 000 lux
on several occasions between 9h00 to 11h00. The peaks are in line with the daily trajectory
of the sun since the sun angle is lower in the summer and winter seasons during sunrise and
sunset. The captured low-angle sun penetration through the windows is shown in Figure 4.5.

From Figure 4.5, the illuminance peaks in spaces 43 and 11 were due to direct sunrays
through the blinds as a result of the low sun angle, as explained above. This resulted in a
maximum illuminance of 12 775.60 lux and 11 711.30 lux in spaces 43 and 11, respectively,
which pose a risk of glare and visual discomfort for the occupants (Kunwar et al., 2020).
According to Kunwar et al. (2020:5), daylight illuminance above 2 000 lux is likely to cause
visual discomfort and glare in direct contact with the human eyes.

53
 Figure 4.5: Sun rays falling on the desk and light sensor in space 11

Offices on the west side of the building were found to have the least daylight illuminance, with
an average of 39.52 lux and 58.86 lux in spaces 49 and 22, respectively. Once again, these
findings are in line with the daily trajectory of the sun. Spaces on the west side only receive
direct sunrays late in the afternoon in the summer season. In the winter season, the spaces
are in the shade throughout the day due to the sunset azimuth angle relative to the orientation
of the building. In addition, the location of the sensor, i.e., the distance of the office desk from
the window, is another contributing factor to the low levels of daylight illuminance observed.

Furthermore, the daylighting in each of the selected offices was analysed in summer and
winter with clear and cloudy sky conditions. Based on the sky conditions assumption as
discussed in Section 4.2.1, GHI and HDI for typical summer clear and cloudy sky days, as well
as the resultant daylight illuminance in the selected offices, are presented in Figure 4.6.

On a typical summer clear sky day, as presented in Figure 4.6 (a), the average GHI between
07h00 and 17h30 was 439.0 W/m2, peaking at 12h00 at 743.10 W/m2. In the same period, the
corresponding average HDI was found to be 45 021.59 lux and peaking at 75 140.0 lux
between 11h00 to 13h00. From Figure 4.6 (c), the average GHI was 85.81 W/m2 between
07h00 and 17h30, with a maximum of 234.50 W/m2 at 14h00. The resultant HDI was 9 160 lux
average with a peak illuminance of 26 490.0 lux at 11h30. As expected, a significant difference
was observed between the clear and cloudy sky GHI.

54
Figure 4.6: Summer season: (a) clear sky day global horizontal irradiance and horizontal daylight illuminance, (b) clear sky day indoor daylight illuminance of
selected offices, (c) cloudy sky day global horizontal irradiance and horizontal daylight illuminance, and (d) cloudy sky day indoor daylight illuminance of
selected offices

55
Clear and cloudy sky GHI differences were also observed in the indoor daylight illuminance of
the offices. As shown in Figure 4.6 (b), spaces 43, 11, and 15 illuminances exceeded the
South African workplace light benchmark of 300 lux. However, this was not the case on the
cloudy sky day, where the illuminance of all the offices was below the 300 lux benchmark. A
synopsis of the selected offices’ illuminance in summer clear and cloudy sky days are
presented in Table 4-2, where the actual daylight illuminance of the selected offices was below
100 lux on a typical cloudy sky day. In contrast, the actual daylight illuminance in spaces 43,
11 and 15 was above 100 lux on a clear sky day. Although there was no indication of daytime
use of electric lights in Figures 4.6 (b) and (c) during the sun’s absence, potential external
(hallway) illuminance was also taken into consideration in the actual daylight illuminance in
spaces 41, 43, 49 and 22.

Further, the sun absence and presence times in Table 4-2 are based on the sun absence and
presence assumptions given in Figure 4.2. The sun absence and presence times in Table 4-
2 are slightly different from Table 4-1 because Table 4-1 is based on the entire monitoring
period, while Table 4-2 focuses on summer season only.

Table 4-2: Typical clear and cloudy sky illuminance levels of the selected offices in the
summer season
Clear sky day
Sun absence (18h00 – 06h00) Sun presence (07h00 – 17h30)
space Avg. Max. Min. Avg. Max. Peak Actual daylight
(lux) (lux) (lux) (lux) (lux) time (lux)
41 10.89 11.80 3.90 22.92 59.10 11h00 12.03
43 19.70 19.70 19.70 489.52 8770.70 10h00 469.82
49 11.80 11.80 11.80 15.75 27.60 10h00 3.95
11 19.70 19.70 19.70 236.16 1257.50 10h00 236.16
15 0 0 0 375.95 1115.0 08h00 375.95
22 11.80 11.80 11.80 63.78 114.30 10h00 51.98
Cloudy sky day
Sun absence (18h00 – 06h00) Sun presence (07h00 – 17h30)
space Avg. Max. Avg. Max. Peak Actual daylight
Min.
(lux) (lux) (lux) (lux) time (lux)
41 11.80 11.80 11.80 11.8 19.70 14h00 0
43 19.70 19.70 19.70 80.99 208.90 14h00 61.29
49 11.80 11.80 11.80 11.80 11.80 - 0
11 19.70 19.70 19.70 40.49 74.9 14h00 20.79
15 0 0 0 67.55 178.0 14h00 67.55
22 11.80 11.80 11.80 22.57 43.4 14h00 10.77

56
From Figure 4.7, relatively low GHI and HDI on the cloudy sky day resulted in a similar poor
indoor daylight illuminance in all the offices, especially in the winter season. Between 07h30
to 17h00, the average cloudy sky day GHI was 47.74 W/m2 with a maximum of 110.6 W/m2 at
11h30. The average HDI was 4 953.0 lux with a maximum of 10 250 lux, peaking several times
of the day. In Figures 4.7 (b) and (d), space 41 and 22 illuminance were unusually higher than
their respective sun absence average illuminances of 10.89 lux and 11.8 lux on a clear sky
day, as shown in Table 4-2. This implies the use of electric lights in the absence and presence
of the sun. On the clear sky day, the electric lights in space 41 were switched on for 18.5
hours, between 00h00 and 18h00, while in space 22, the lights were on for 24 hours. On the
cloudy sky day, the electric lights in space 41 were observed to be on for 3.5 hours, from
16h00 to 19h00 and 11 hours, between 16h00 and 19h00 in space 22. Table 4-3 presents a
synopsis of typical winter clear and cloudy sky days illuminance of the selected offices.

57
Figure 4.7: Winter season: (a) clear sky day global horizontal irradiance and horizontal daylight illuminance, (b) clear sky day indoor daylight illuminance of
selected offices, (c) cloudy sky day global horizontal irradiance and horizontal daylight illuminance, and (d) cloudy sky day indoor daylight illuminance of
selected offices

58
Table 4-3: Typical clear and cloudy illuminance of the selected office in the winter season
Clear sky day
Sun absence (17h30 – 06h30) Sun presence (07h00 – 17h00)
Space Avg.
Avg. (lux) Max. (lux) Min. Max. (lux) Peak time Avg. (lux)
(lux)
41 71.37 122.20 3.90 137.96 185.3 13h00 66.59
43 19.70 19.70 19.70 109.06 500.60 9h30 89.36
49 7.41 11.80 11.80 10.67 19.70 10h00 3.26
11 19.70 19.70 19.70 154.41 587.30 10h30 154.41
15 0 0 0 159.28 560.0 12h30 159.28
22 99.40 114.3 90.70 149.61 193.20 11h00 50.21
Cloudy sky day
Sun absence (17h30 – 06h30) Sun presence (07h00 – 17h00)
Space Avg.
Avg. (lux) Max. (lux) Min. Max. (lux) Peak time Avg. (lux)
(lux)
41 28.14 130.10 11.80 28.36 122.20 17h00 0
43 19.70 19.70 19.70 37.07 59.10 11h00 37.07
49 5.59 11.8 3.90 6.27 11.80 15h00 6.27
11 19.7 19.70 19.70 28.79 35.50 10h00 28.79
15 0 0 0 34.55 77.0 11h30 34.55
22 30.67 122.20 11.80 108.02 130.10 14h30 77.35

The average actual daylight illuminance of all offices on a cloudy sky day was below 100 lux,
and only spaces 11 and 15 daylight illuminances were above 100 lux on a typical winter clear
sky day. The number of hours when daylight illuminance was above the 300 lux benchmark
was used to determine the daylight performance of the selected offices. In other words, based
on an on/off approach for electric lights, offices with illuminance greater than 300 lux in the
presence of the sun can sufficiently utilise daylight for office tasks without the need for electric
lights. A similar principle was adopted by Lim et al. (2017) to determine the automated on/off
daylight-linked energy savings in a green design office building in Malaysia. The automated
on/off daylight-linked principle was used in this experiment to determine the potential daylight
energy savings of the selected offices.

4.2.3 As-is daylighting energy performance and potential

Although daylight-linked sensors were not installed in the selected offices, the automated
on/off daylight-linked principle of operation was used to derive the experimental “as-is” daylight
energy performance of the offices. As indicated in Chapter 2, the on/off daylight-linked
principle is based on the automated operation of electric lights using pre-set illuminance. The
electric lights are automatically switched off by the daylight-linked sensor when indoor
illuminance is equal to or above the pre-set illuminance and on when the illuminance is below
(Lim et al., 2017). The selected offices’ light energy consumption and potential daylight energy
savings were computed using Equation (1) and appendix A-1 and A-2. Figure 4.8 shows

59
typical summer clear sky lighting and daylight-linked energy performances for the selected
offices.

In Figure 4.8, the baseline represents the “as-is” lighting profile of the selected offices without
the use of a daylight-linked sensor, while daylight-linked is the lighting profile based on the
principle of on/off daylight-linked operation. Considering that the illuminances of all the offices
were below the 300 lux benchmark on cloudy sky days (see Figures 4.6 and 4.7), their lighting
energy is equal to the baseline light energy in Figure 4.8. This is also applicable to space 22,
while space 49 has a distinctive profile that will be covered later in this section.

From appendix A-1 and A-2, the total lights wattage of the selected offices can be grouped
into 420 W, and 140 W. Spaces 41 and 49 have a total lights-wattage of 420 W each, while
spaces 43, 11, 15, and 22 have 140 W each.

60
Figure 4.8: Typical summer season clear sky daylight energy consumption and daylight energy savings

61
In terms of burning hours, 08h00 to 17h00 is the weekday time frame for classes and
laboratory activities, according to the University of Fort Hare Faculty of Science and
Agriculture 2018 prospectus. However, the measured indoor illuminance of the selected
offices (see Figure 4.2 and Table 4-1) shows activities in some of the offices beyond the
above-mentioned time frame. It was observed that space 41 electric lights were switched on
57% (47 days) of the entire monitoring period. Space 22 had the second-highest duration of
electric lights usage outside the working hours with 11% (9 days). Considering the Faculty
time frame and measured indoor illuminance, the following assumptions were made to
establish burning hours of electric lights in the selected offices.

● All electric lights in a given space were operated by a control switch and were switched
on whenever space was occupied.
● The electric lights in space 41 were switched on 24 hours of the weekdays and
weekends.
● The electric lights in spaces 43, 49, 11, 15, and 22 were switched on between 08h00
and 20h00 on weekdays and switched on from 10h00 to 19h00 on weekends.
● The lighting operation pattern is the same in the summer and winter seasons.

A summary of the lighting energy performance of the selected offices is presented in

Table 4-4.

Table 4-4: Summer clear sky day lighting energy performance of the selected offices
Baseline Daylight-linked
Energy Daylight
Space Energy (kWh/day) Saving (%)
Demand (W) (kWh/day) period
Wkday Wkend (h) Wkday Wkend Wkday Wkend
41 420 10.08 10.08 0 0 0 0 0
43 140 1.89 1.19 0.5 1.82 1.19 4 0
49 420 5.67 3.57 0 0 0 0 0
11 140 1.89 1.19 3.0 1.47 1.05 22 12
15 140 1.89 1.19 4.5 1.26 0.91 33 24
22 140 1.89 1.19 0 0 0 0 0

As indicated in Table 4-4, space 41 had the highest lighting energy consumption due to its
total light-wattage and assumed lights burning hours. In contrast, space 15 was found to have
the most daylight-linked energy saving potential, with 33% and 24% on weekdays and
weekends, respectively. Typical winter clear sky day lighting energy performance was similarly
computed using Equation (1) and Appendix A-1 and A-2 and presented in Figure 4.9.

62
Figure 4.9: Typical winter clear sky day lights energy consumption and daylight energy savings

63
Similar approaches and assumptions were adopted in computing the winter clear sky day
lighting energy of the selected offices. The baseline energy was the same, with an average of
10.08 kWh/day for space 41, 1.69 kWh/day for spaces 43, 11, 15, and 22, and 5.58 kWh/day
for space 49. Space 41 was still the most underperforming, with the highest lighting energy
consumption and no daylight-linked energy saving potential. The daylight-linked energy
consumption for spaces 43, 11, and 15 was 1.61 kWh/day each on weekdays, saving 15%.
On weekends, the daylight-linked energy consumption for spaces 43 and 11 was 1.12
kWh/day with 41% savings, while 0.91 kWh/day was obtained in space 15, thus amounting to
52% energy savings. Further, the energy charges of the MEGA FLEX tariff structure used in
computing the selected spaces lighting energy costs are presented in Table 4-5 (Eskom,
2021).

Table 4-5: MEGA FLEX Nightsave Urban Small tariff energy charges (Eskom, 2021)
High demand season (Jun- Low demand season (Sep-
Transmission Aug) VAT incl.(c/kWh) May) VAT incl. (c/kWh)
Voltage
Zone Off- Off-
Peak Standard Peak Standard
peak peak
< 500 V 479.96 146.03 79.74 157.17 108.45 69.14
≥ 500 V &
472.43 143.12 77.73 154.09 106.08 67.3
< 66 kV
≤ 300 km
≥ 66 kV &
457.47 138.58 75.27 149.25 102.7 65.18
≤ 132 kV
≥ 132 kV 431.15 130.59 70.93 140.68 96.8 61.42

From Table 4-5, the MEGA FLEX energy charges vary with season. June to August, which is
considered the winter (high demand) season, has the highest energy charges, while
September to May represents the summer (low demand) season and has the least energy
charges. The energy charges are further classified according to time of use (TOU). The TOU
include peak, which consists of the morning peak, 07h00 to 09h30 and the evening peak,
18h00 to 19h30. Also, the standard TOU is made up of morning standard, 6h00 to 6h30,
midday standard, 10h00 to 17h30, and evening standard, 20h00 to 21h30. Lastly, the off-peak
TOU includes morning off-peak, 0h00 to 5h30 and evening off-peak, 22h00 to 23h30. It is
worth noting that the tariff time frames are given in 30 minutes intervals, so a minute after a
given time frame, the nearest tariff applies. Further, the least transmission zone was adopted
in the research since the nearest sub-station is 100 km away from the University. Considering
the above energy-charge factors, Equation (2) and Figure 4.8 combined with Table 4-5 were
used to deduce the typical summer clear sky day operational lighting energy cost of the
selected offices, which are presented in Figure 4.10.

64
 Figure 4.10: Summer season clear sky day light energy cost of the selected offices

The low demand energy charges were adopted in computing summer clear sky day lighting
energy cost given in Figure 4.10. The data indicate that Space 41 incurred the highest lighting
energy cost, with an average of R2.5 per hour, amounting to a daily cumulative cost of R10.63.
The daily cumulative lighting energy cost for spaces 43, 11, 15, and 22 was R 2.37 on
weekdays and R1.39 on weekends, while space 49 lighting energy cost amounted to
R7.17/day on weekdays and R4.18/day on weekends. However, spaces 43, 11, and 15 have
the potential to utilise daylight to reduce their respective lighting energy cost by 4%, 34%, and
57% per week. Similarly, Equation (2) and Figure 4.9, together with Table 4-5, were used to
derive the lighting energy cost of the selected offices on a typical winter clear sky day, as
presented in Figure 4.11.

65
 Figure 4.11: Winter season clear sky day light energy cost of the selected offices

As expected, the winter season lighting energy cost was higher by approximately 50% due to
the high demand (winter) charges. The average daily cumulative lighting energy cost for space
41 was R 19.51, while space 49 had a total daily lighting energy cost of R 15.29 on weekdays
and R 7.32 on weekends. The baseline lighting energy cost for spaces 43, 11, 15 and 22 was
R 5.10 and R 2.44 on weekdays and weekends, respectively. The lighting energy cost savings
were 17% per week in spaces 43 and 11, as well as in space 15 by 25% per week.

4.3 Conclusions

This Chapter presents data, analyses and sub-findings on the experimental and simulation
“as-is” daylighting and energy performance of the case study building. The impact of the global
horizontal irradiance (GHI) and horizontal daylight illuminance (HDI) was demonstrated
through the experimental-based evaluation of the building. Cloudy sky days in summer and
winter seasons showed the least GHI and HDI, which correlated with poor indoor illuminance
below the benchmark of 300 lux at the workplace level. In addition, building orientation relative
to the daily traverse of the sun also influenced effective daylighting indoors such that offices
on the west side of the building had the least daylight illuminance on typical summer and winter
clear sky days, with their average illuminance below 100 lux. On the other hand, the average
illuminance in the east-facing offices was up to 469.82 lux, with the maximum daylight
illuminance observed in the winter season. Effective daylighting and daylight-linked energy-

66
saving potential of the offices were evaluated against the daylight illuminance versus the South
African workplace benchmark of 300 lux. The evaluation shows that only three out of the six
offices sampled had effective daylight and hence a potential to utilise daylight-linked control
for energy savings.

67
 5CHAPTER 5

Simulation and Optimisation of the Building Daylight and Lighting Energy

Performances

5.1 Introduction

Data and analyses in the previous Chapter indicate that electric lighting was the second-
highest energy consumer, mainly due to the high number of lights in the building and the
absence of automated control. All the lights were scheduled to operate whenever a given
space was occupied, regardless of the indoor daylight illuminance per the common practice
in commercial buildings (Escrivá-Escrivá, 2011). However, daylight-linked sensors based on
ambient light levels can be used to aid the operation of lights to minimise energy consumption
(Alrubaih et al., 2013). According to Atif and Galasiu (2003), the common operations of
daylight-linked sensors are automated on/off and continuous dimming. Irrespective of the
operation, daylight-linked sensors are usually installed at the ceiling facing downward (Lim et
al., 2017). Other lighting interventions include occupancy-linked and time-based light sensors.
The operating principle of occupancy-linked sensors was covered in Chapter 2 (see Section
2.3.1), but it was not considered in the simulation for building lighting energy optimisation.

This Chapter covers daylight interventions for the building lighting and energy performance
optimisation and reduction of the overall energy use intensity (EUI) as well as economic and
environmental impacts. To this effect, daylight-linked sensors, as explained above, were
adopted in the simulation for building energy optimisation. The building daylight distribution
and performance analysis based on indoor luminance and illuminance of various spaces in
the building was used to kick-start the Chapter.

5.2 Simulation-based Analysis of Daylighting and Energy Performance Analysis of

the Building

The experimental section covers the daylight and lighting energy performance of the building
with respect to selected offices under normal operating conditions. The effect of the blinds and
the operation of the electric lights on the illuminance of the selected offices were
demonstrated. In contrast, the simulation-based analysis covers the “as-is” daylighting and
whole-building energy performance. The simulation daylight evaluation of the building is
mostly based on design conditions. In other words, the effect of the blinds and electric lights
were not considered in the simulation. Further, the building daylighting and energy simulation

68
was guided by the local sun path and solar radiation analysis of the building, as covered in the
next section.

5.2.1 Sun path and solar distribution for the building

As indicated in Section 4.2, the seasonal daily trajectory and altitude of the sun are critical
factors of effective daylighting in a building. A sun path diagram was therefore generated from
satellite-based weather data (see Figure 5.1) and applied to evaluate the seasonal daily
traverse of the sun relative to the building.

Figure 5.1: Sun path diagram for the latitude of the building

As highlighted in Section 3.3.2, the nearest available local weather data for the building in the
IESVE database was from Bisho town. In Figure 5.1, 21 June and 21 December are
respectively winter and summer solstice, 21 March and 21 September are the equinoxes,
while 01 January to 31 December was selected for annual simulation of the sun path. During
the summer solstice, the sun rises at 05h00 due south-east (118.28°), peaks at 12h00 at an
altitude of 80.0° and sets at 19h00 due south-west (241.70°). A shorter daytime was observed
in the winter solstice as the sun rises at 07h00 due north-east (61.70°) and sets at 17h00 due
north-west (298.28°). Hence, the building winter season daylight time is approximately 4 hours
shorter. In addition, the sun rises and sets approximately due east (90°) and west (270°) at

69
06h00 and 18h00, respectively, in the equinox. A synopsis of the annual sun altitude at
different times of the month is given in Table 5-1.

Besides the sun azimuth, solar altitude is another important parameter in terms of the
penetration of the sun rays into a building. Theoretically, the roof overhang or eaves prevents
the summer high altitude sun rays from penetrating through the windows into the building while
the winter low altitude sun rays can penetrate deeper (Overen, Meyer and Makaka, 2019).
The solar altitude angles in Table 5-1 were simulated using the 15th of each month as the
design day (typical day). September to May is also considered the summer season, while June
to August is the winter season. The average summer sun altitude was 36.82° with a peak of
80.31 in December, while an average altitude of 23.34° was observed in winter with a peak of
43.18° in August (as one approaches the summer period). The average seasonal sun altitude
corresponded with Overen et al.’s (2019) findings and justified the difference in seasonal solar
penetration angles in the building as shown in Figure 2.2.

A solar shading analysis of the building was conducted, and the findings are presented in
Appendix B-1 and B-2. In the solar shading tables, the dark grey sections indicate the times
the sun is below the horizon, while the values in the shaded-blue and shaded-yellow are the
percentage area of the floor or wall of a particular space that is receiving solar radiation at a
given time of the month. The shaded-blue tables contain values for the floor, while the wall’s
values are given in the shaded-yellow tables. The floor readings indicate internal solar
radiation due to penetration of the sun rays, while the wall readings were due to external solar
radiation falling on the outer surface of the walls. As shown in Appendix B-1 and B-2, the
percentage area of floor and walls receiving solar radiation varies with time, orientation, and
season. A virtual illustration of the solar exposure of the building in four different orientations
on a typical summer day is presented in Figure 5.2.

In Figure 5.2. the surface of the entire building was divided into a 10 m2 imaginary grid for the
hourly solar exposure analysis. A scale of 0 to 12 hours was selected to accommodate the
summer and winter seasons on one scale. From Figure 5.2, only the roof was found to have
solar exposure of 12 hours and above. This is followed by the top floor north-facing walls
(spaces 14, 16, and 18) with approximately 8 hours of solar exposure. The west-facing walls
of space 39 (ground floor) and some portion of space 23 (top floor) were observed to have no
solar exposure throughout the day. As indicated in Figures 3.2 and 3.11, the poor solar
exposure of the above-mentioned walls is due to the orientation of the building and trees. As
shown in the south-east orientation in Figure 5.2, the shading effect of adjacent buildings was
observed on the ground floor walls of spaces 41, 42, and 43.

70
Table 5-1: Annual solar altitude of the building

71
Figure 5.2: Simulated hourly solar exposure of the building on a typical summer day

72
Figure 5.3: Simulated hourly solar exposure of the building on a typical winter day

73
On the north-west elevation, space 49 walls were found to be affected by the adjacent building
and trees, while self-shading resulted in the relatively low solar exposure of walls for space
46. Figure 5.3 presents the hourly solar exposure of the building on a typical winter day.

The maximum solar exposure in Figure 5.3 was approximately 10 hours. The south-facing
walls were completely in the shade with zero-hours of solar exposure throughout the day.
Spaces 41, 42, and 43 solar exposure hours were found to drop from approximately 6 hours
in summer to 2 hours in winter. The west walls, specifically the ground floor, experienced a
similar decrease in solar exposure hours. As explained earlier, the relatively low solar
exposure hours observed in the winter season are a result of the short winter sun hours. The
angles of sunrise and sunset relative to the building orientation and the adjacent structure on
the east and west side of the building also contributed to the relatively low solar exposure
hours. As discussed further in the next section, the solar exposure hours of the walls are
equivalent to the solar exposure of the respective windows on the walls, and this influences
the available daylight in the spaces.

5.2.2 Daylight distribution and performance of the building

Given that the building solar exposure hours covered in the previous section do not represent
the ambient or indoor light levels, luminance and illuminance were used in this section to
analyse the ambient and indoor daylight levels relative to the building orientation and adjacent
structures. Figure 5.4 therefore shows the building luminance for various orientations on
typical summer and winter days.

Figure 5.4: Simulated typical summer and winter days outdoor ambient luminance for
various orientations

Luminance is simply the brightness of light in a given space or area, thus, Figure 5.4 illustrates
the brightness of sunlight on the building walls. The windows were not considered since the
majority of sunlight falling on the window (glass) is transmitted into the building. However, it

74
can be assumed that the amount of sunlight falling on the walls is equivalent to the amount
falling on windows located on that wall. In Figure 5.4, the impact of the adjacent structure on
the building was evident in the winter season at the north-west and north-east elevations. In
both elevations, the adjacent building was shading more than half of the west and east-facing
walls. The average luminance of the east-facing wall portion in the shade was 350 cd/m2, while
the non-shaded portion was 5 700 cd/m2. The average luminance of the east wall in summer
was 5 500 cd/m2. On the west-facing walls, shading due to the adjacent structure reduces the
average luminance from 1 600 cd/m2 to 240 cd/m2. However, the average wall luminance was
7 400 cd/m2.

The luminance of the south-west and south-east walls is a factor of seasonal time of sunrise
and sunset relative to the building orientation. Nonetheless, the average luminance of 5 600
cd/m2 in summer and 6 500 cd/m2 in winter was observed on the south-east facing walls. The
south-facing walls were found to be partially shaded in summer, resulting in an average
luminance of 1 500 cd/m2. In winter, the south-facing walls were completely shaded with an
average luminance of 500 cd/m2. For comparison purposes, sunlight on a sandy beach with
direct contact with the human eyes is 28 600 cd/m2, while a 40 W bulb is 3 068 cd/m2
(Szokolay, 2008). The author (ibid.) alludes that lights with 25 000 cd/m2 may result in glare
when directly sensed by human eyes.

Furthermore, summer and winter days indoor luminance of various spaces on the ground floor
of the building were also simulated and presented in Figures 5.5 and 5.6.

75
Figure 5.5: Simulated indoor luminance of the ground floor spaces on a typical summer day

76
Figure 5.6: Simulated indoor luminance of the ground floor spaces on a typical winter day

77
In Figures 5.5 and 5.6, 10 spaces in the building (including offices, laboratories and a lecture
hall) were sampled. Spaces 37, 39, 46, 48, and 49 are located on the west side, space 35 is
south facing, while spaces 41, 42, 43, and 45 are located on the east side. As in Figure 5.4,
for spaces on the west side, indoor luminance was simulated at 16h30, while the luminance
of spaces on the east side was simulated at 10h00. This corresponds to the periods of
maximum ambient luminance at the building’s west and east sides, as shown in Figure 5.4. A
scale of 0 to 1000 cd/m2 was adopted in Figures 5.5 and 5.6. Once again, the effect of the
building orientation relative to the time and angles of sunrise and sunset were evident in
Figures 5.5 and 5.6. Besides spaces 43 and 45, the average luminance of the spaces in
summer was higher compared to levels in winter. Space 35 had the least luminance of 16.12
cd/m2 in summer, while the lowest luminance of 2.96 cd/m2 in winter was observed in space
49. In contrast, space 49 had the highest luminance of 216.96 cd/m2 in summer, alongside
space 46 with 77.46 cd/m2 in winter. It is also evident in Figures 5.5 and 5.6 that the luminance
of the spaces decreases as the distance from the windows increases. This was also noticed
in spaces 46, 48, 49, 41, 42, 43, and 45 in the summer season. The highlighted spaces all
have windows on one external wall such that daylight is received from one direction, which
thus creates an uneven distribution of daylight indoors.

In addition to the luminance levels of the spaces, possible instances of visual discomfort due
to glare were also analysed Based on Guth Visual Comfort Probability (GVCP) and Daylight
Glare Index (DGI). Dolnikova and Katunsky (2019) define GVCP as the percentage of
occupants who would be visually comfortable facing a specified direction relative to the light
source in a space. The GVCP ranges from 0% to 100%, whereby 0% indicates that none of
the occupants in the space would find their view path visually comfortable, while 100% means
all the occupants would be visually comfortable with their view path. The author (ibid.:5)
alludes that 70% GVCP is the acceptable benchmark, but computer tasks require 80%. In
contrast, DGI is the degree of visual discomfort of an occupant relative to the view path and
the light origin (Bellia, Cesarano, Iuliano and Spada, 2008). DGI ranges from ≤16, which is
considered “just imperceptible”, to >28, which is considered “intolerable” (Detsi, Manolitsis,
Atsonios, Mandilaras and Founti, 2020). Based on Figures 5.5 and 5.6, the simulated GVCP
and DGI of the ground floor spaces are summarised in Appendix C-1.

Similarly, the indoor luminance of space for the top floor spaces on typical summer and winter
days are presented in Figures 5.7 and 5.8, respectively. The spaces sampled consist of four
laboratories, five offices, and a lecture hall. Spaces 18, 22, and 23 are located on the west
side of the building, while spaces 7, 11, 14, and 15 are located on the east side and spaces 3

78
and 5 are located on the south, with space 16 located on the north side. The highest luminance
of 950.67 cd/m2 was observed in space 16 in winter, while space 3 had the least luminance of
4.60 cd/m2 in the same season. In the summer season, the same space (space 3) had the
least luminance of 7.49 cd/m2, while the maximum luminance of 151.15 cd/m2 was observed
in space 22. Also, the decrement of luminance away from the windows was observed in
spaces 18, 11, 15, and 14 for both seasons. Based on Figures 5.7 and 5.8, the simulated
GVCP and DGI analysis of the top floor spaces of the building are summarised in Appendix
C-2.

79
Figure 5.7: Simulated indoor luminance of the top floor spaces on a typical summer day

80
Figure 5.8: Simulated indoor luminance of the top floor spaces on a typical winter day

81
A comparison of luminance between ground floor and top floor spaces was conducted and
presented in Table 5-2. In general, spaces to the west side of the building had the highest
luminance levels with an average of 104.32 cd/m2 in summer, while spaces to the south had
the least luminance levels with an average of 5.18 cd/m2 in winter. These do not include the
north-facing office, which has an average luminance of 950 cd/m2. In terms of storeys, the top
floor spaces had the highest luminance, with an average of 47.39 cd/m2 in summer and 123.96
cd/m2 in winter. The average luminance levels for spaces on the ground floor was 75.39 cd/m2
and 22.12 cd/m2 in the summer and winter, respectively.

Table 5-2: Comparison of ground and top floor spaces luminance

Summer Winter
Orientation Ground floor Top floor Ground floor Top floor
(cd/m2) (cd/m2) (cd/m2) (cd/m2)
South 16.12 8.56 8.39 5.18
East 54.05 42.17 42.77 49.14
North 0.00 51.22 0.00 950.67
West 104.32 78.97 8.35 27.35

Seasonally, the highest average luminance of 73.04 cd/m2 is experienced in winter, which was
11.65 cd/m2 higher than the average luminance in summer. Nonetheless, spaces on the east
side of the building experienced the most stable luminance, with an average of 47.03 cd/m2
for both seasons and floors.

Theoretically, the luminance of a space is a factor of the amount of light, reflectivity, and the
colour of materials. As indicated in Figures 5.5 to 5.8, bright coloured (white) surfaces tend to
have higher luminance compared to dark (black) coloured surfaces. Indoor illuminance (actual
levels of light in a space) was therefore analysed to categorically characterise daylight
distribution. In line with Kruisselbrink et al. (2018), this was based on an imaginary grid of 1 m
x 0.5 m and, as indicated by the red border lines, area of interest (AOI), which is 0.5 m away
from the walls and windows of the spaces, was adopted. In addition, the building illuminance
was simulated with a work plane of 0.85 m.

Further, Figures 5.9 and 5.10 show the indoor illuminance distribution for ground floor spaces
on typical summer and winter days, respectively. The green shades indicate areas in the AOI
with illuminance lesser than 300 lux.

82
Figure 5.9: Simulated indoor illuminance of the ground floor spaces on a typical summer day

83
Figure 5.10: Simulated indoor illuminance of the ground floor spaces on a typical winter day

84
From Figure 5.9, spaces 41, 42, 43, 46, 48, and 49 experienced an average illuminance above
2 000 lux, while spaces 35, 37 and 39 had average illuminances below 1 000 lux. Space 39
was observed to have the least illuminance, with an average of 23.01 lux. The maximum
average illuminance of the spaces was 6 347.04 lux and was observed in space 49.
Comparing both seasons, the average illuminance in winter (509.90 lux) was 84% less than
in summer (3 244.88 lux). From Figure 5.10, average illuminances above 1 000 lux were
observed only in spaces 43 and 45. The maximum average illuminance observed was
2 237.69 lux for space 45, while space 39 had the lowest average of 35.30 lux. A statistical
summary of Figures 5.9 and 5.10 is presented in Table 5-3.

Table 5-3: Statistical description of the building ground floor spaces illuminance
Minimum Maximum Average AOI < 300
Space Season Uniformity Diversity
(Lux) (Lux) (Lux) lux (%)
Summer 12.87 558.11 82.32 0.156 0.023 99.38
35
Winter 7.12 299.55 47.76 0.149 0.024 100
Summer 5.14 35410.19 796.73 0.006 0 84.15
37
Winter 0.94 4048.76 75.24 0.012 0 97.78
Summer 0.43 2976.64 23.01 0.019 0 99.88
39
Winter 1.7 327.62 35.3 0.048 0.005 99.69
Summer 44.02 41713.18 4422.32 0.01 0.001 32.29
41
Winter 6.34 21647.99 631.33 0.01 0 78.58
Summer 24.44 42084.58 4494.54 0.005 0.001 34.69
42
Winter 6.54 743.92 138.09 0.047 0.009 88.51
Summer 12.93 41700.92 3644.55 0.004 0 49.0
43
Winter 49.37 31038.13 1734.09 0.028 0.002 22.29
Summer 3.13 42031.59 4058.06 0.001 0 61.7
45
Winter 4.71 45551.25 2237.69 0.002 0 45.88
Summer 87.54 45401.21 3564.94 0.025 0.002 8.34
46
Winter 2.14 350.5 60.06 0.036 0.006 99.79
Summer 49.56 36809.83 5015.29 0.01 0.001 6.77
48
Winter 2.28 304.87 73.56 0.031 0.007 99.99
Summer 109.92 30504.87 6347.02 0.017 0.004 0.31
49
Winter 1.41 253.02 65.92 0.021 0.006 100

In the context of daylight, uniformity is the quality of daylight in the AOI, and it is mathematically
defined as the ratio of the minimum to the average illuminance of the AOI (Kruisselbrink et al.,
2018). Diversity is the spread of daylight in the AOI, and it is mathematically expressed as the
ratio of the maximum to the minimum illuminance of the AOI (Elsiana, Soehartono and
Kristanto, 2020). In Table 5-3, relatively low daylight uniformity and diversity were observed in
the spaces compared to the findings of Freewan and Al Dalala (2020). Unlike the case study
building in this research, the authors (ibid.) assess the daylighting performance of passive

85
solar-designed classrooms building in a Jordan University. The classrooms building daylight
illuminance uniformity was found to be in the range of 0.15 to 0.66. Although the uniformity of
both buildings was not expected to be equal, given the difference in their design, the
comparison creates a perspective of modern building daylight illuminance uniformity and the
impact of design on daylight distribution. Besides space 35, the illuminance uniformity of the
spaces was below 0.1. As discussed in detail in Chapter 2, Section 2.4.1, the low uniformity
of the spaces is due to the side-lighting strategy adopted in the building. Bradshaw (2006)
emphasises that side-lighting is associated with poor indoor light quality due to possible
blockage by adjacent structures outdoors and limited depth of diffusion relative to the position
of the vertical windows. The furniture in the spaces, together with the side-lighting, also creates
an uneven distribution of light in the space, thereby reducing illuminance uniformity and
diversity.

Space 35 had higher uniformity and diversity of illuminances due to the lack of direct
penetration of sun rays. Due to the same reason, space 35 was found to have the highest
percentage AOI with illuminance lesser than 300 lux. In other words, the illuminances of
99.38% and 100% of space 35’s AOI were less than 300 lux in summer and winter,
respectively. In contrast, only 0.31% of space 49’s AOI was found to have illuminance lesser
than 300 lux in summer but shares the highest percentage AOI with illuminance less than 300
lux in winter. Moreover, space 43 had the most consistent daylight illuminance in both
seasons, with only 49.0% and 22.29% of its AOI at illuminances less than 300 lux in summer
and winter, respectively. Further, the indoor illuminance distribution of the top floor spaces of
the building on typical summer and winter days is presented in Figures 5.11 and 5.12.

In Figure 5.11, 50% of the spaces (spaces 11, 14, 15, 18 and 22) have average illuminances
greater than 2 000 lux. Space 22 had the highest average illuminance of 6 411.59 lux, while
the least average illuminance of 69.29 lux was observed in space 5. The average illuminance
of spaces 5, 3, 7, 16, and 23 was below 1 000 lux. A relatively smaller difference in illuminance
(at 57%) was observed in the top floor spaces in summer and winter.

86
Figure 5.11: Simulated indoor illuminance of the top floor spaces on a typical summer day

87
Figure 5.12: Simulated indoor illuminance of the top floor spaces on a typical winter day

88
In Figures 5.11 and 5.12, the average illuminance of spaces 11 and 14 were above 2 000 lux,
while that of spaces 15 and 18 was above 1 000 lux but below 2 000 lux, and the illuminance
of the other spaces was below 1 000 lux. Spaces 14 and 5 had the highest and least average
illuminances of 2 507.81 lux and 38.37 lux, respectively. Table 5-4 contain a statistical
description of Figures 5.11 and 5.12.

Table 5-4: Statistical description of the building’s top floor spaces illuminance
Minimum Maximum Average AOI < 300
Space Season Uniformity Diversity
(Lux) (Lux) (Lux) lux (%)
Summer 2.06 564.68 87.59 0.024 0.004 98
3
Winter 1.05 432.56 51.24 0.021 0.002 99.99
Summer 5.9 282.64 69.29 0.085 0.021 100
5
Winter 2.2 157.79 38.37 0.057 0.014 100
Summer 1.22 54804.93 807.47 0.002 0 90
7
Winter 1.2 37544.01 842.8 0.001 0 85
Summer 69.4 42017.08 3651.9 0.019 0.002 49.87
11
Winter 101.95 31573.65 2507.81 0.041 0.003 38.07
Summer 0.86 41720.25 4872.37 0 0 36.99
14
Winter 2.22 26260.58 2904.01 0.001 0 33.43
Summer 0.7 43713.34 3980.02 0 0 64.23
15
Winter 5.14 33727.72 1712.28 0.003 0 58.77
Summer 0 831.12 125.41 0 0 91.34
16
Winter 0 28026.93 1779.03 0 0 26.76
Summer 16.51 36577.68 3789.54 0.004 0 32.58
18
Winter 3.46 4558.46 146.23 0.024 0.001 95
Summer 0.07 27706.79 6411.59 0 0 6.07
22
Winter 0.06 3785.92 362.05 0 0 52.18
Summer 0.37 42787.05 596.06 0.001 0 90.52
23
Winter 0.09 2469.29 62.64 0.001 0 98.07

As evident in Table 3-4, a relatively low illuminance uniformity and diversity were observed for
top floor spaces where only spaces 3 and 5 (south-facing spaces) where illuminance
uniformity and diversity were greater than 0.01 in both seasons. As indicated earlier, this can
be attributed to the absence of direct sun rays in south-facing spaces. Both south-facing
spaces recorded the highest percentage AOI with illuminance lesser than 300 lux. Although
space 22 had the least percentage AOI of 6.07%, with illuminance less than 300 lux. Space
14 was found to have the most consistent illuminance distribution in both seasons.

5.3 Whole-building (as-is) Energy Performance

This section covers the whole-building energy performance of the case study building but
mainly focuses on the energy use intensity (EUI). In this regard, only electric lighting and

89
energy-use of appliances in conditioned spaces were considered in the simulation. In the
context of building energy performance, conditioned spaces or zones are areas in a building
that are thermally conditioned mechanically or electrically for thermal comfort. Conditioned
spaces therefore exclude hallways, toilet/bathroom, storage, and garage. The lights and
appliance energy-use simulation were based on Appendix A-1 and A-2 together with the
University of Fort Hare Faculty of Science and Agriculture 2018 prospectus. The building
energy performance simulation was initiated by creating the building lighting and appliances’
demand profiles based on the academic schedule in the prospectus. Table 5-5 contains the
major schedules of the building that were adopted in the simulation.

Table 5-5: Schedule for energy performance simulation of the building

Schedule/profile Daily Weekly Annual
Lecturers 08h00 – 16h30 Mon. – Fri. 01 Jan – 14 Jan Holiday
15 January – 14
08h00 – 16h30 Mon – Fri. In session
June.
Lectures/researchers

Staff
10h00 – 19h00 Sat. – Sun. 15 Jun – 14 Jul Exam/holiday
15 Jul – 15 Dec In session
Cleaners 07h00 – 15h00 Mon. – Fri 16 December –
Exams/holiday
31 December.
Postgraduate Tue, Wed, & 01 January – 31
15h00 – 18h00 Holiday
classes Fri. January.
Undergraduate 01 February –
08h00 – 13h30 Mon – Fri. In session
classes 31 May
First-year Mechanics 01 June – 15
14h00 – 17h30 Mon. – Fri. Exam/holiday
Students

practical July
First-year electrical
08h00 – 17h30 Mon. – Fri. 16 July – 15
practical In session
October.
Optics practical 14h00 -16h30 Tue. & Thur.
Second-year
14h00 – 17h30 Tue. & Thur. 16 October – 31
practical Exam/holiday
December.
Third year practical 14h00 – 17h30 Tue. & Fri.

As indicated in Table 5-5, the energy schedules of the building consist of activities during
weekdays and weekends. From Figure 4.7, it was observed that some of the offices operate
until late (20h00) hours of the weekdays and on weekends as well. This was used to guide
the development of the building energy usage schedules. From Table 5-5, the daily hours of
occupancy were used to develop the weekly schedules of all conditioned spaces in the
building, which were then integrated into an annual schedule, taking the outlined holidays into
consideration. It is worth noting that only major academic breaks such as mid-year and end-
of-year breaks were considered. The annual schedule in Table 5-5 was also used to create

90
the lights and appliances’ schedules in the conditioned spaces. It was assumed that during
the occupancy period, all the lights and appliances were switched on. However, appliances in
this context refer to personal desktops and laptop computers. The operation of other
appliances, such as electric kettle, telephone set, electric heater, fan, and air conditioner, were
assigned different operating profiles. Figures 5.13 and 5.14 show the average simulated
summer and winter demands of electric lights and appliances.

In Figures 5.13 and 5.14, interior lighting represents all electric lights on the ground and top
floor conditioned spaces. The receptacle equipment consists of desktop computers, laptops,
printers, telephone sets and laboratory apparatus. As indicated earlier, desktop computers,
laptops, and speakers share the same operating profile as the lights in their respective spaces.
The printers were scheduled to operate at 10% capacity from 08h00 to 16h30 for spaces only
occupied during the weekdays and from 08h00 to 20h00 for those spaces occupied on
weekdays and weekends. The operating profile of the printers is due to the intermittent use of
printers in a typical office setting. The operation of the laboratory apparatus was scheduled
according to their respective time frames, as outlined in Table 5-5. Further, the power demand
of electric kettles is represented by cooking, while other process indicates the demand for
electric heaters. Other process and interior local fan (a combination of fans and air
conditioners demands) were scheduled based on the indoor air temperature of their respective
spaces. Specifically, fans and air conditioners were scheduled to turn on when the indoor air
temperature was above 24°C, while electric heaters were only switched on when the air
temperature was below 20°C. The indoor air temperature threshold is based on the South
African Bureau of Standards (2011) indoor air temperature setpoint recommendation.

Typical summer weekdays average demand of the interior fan was set at 1.88 kW, while the
average demand of the electric heaters was set at 0 kW. On winter weekdays, the interior fan
average demand was reduced by 1.72 kW, while the average demand for electric heaters was
increased by 0.46 kW, as indicated by other process in Figures 5.13 and 5.14. In both seasons,
all other appliances maintained a constant average demand of 2.50 kW for interior lighting,
3.60 kW for receptacle equipment, 0.12 kW for cooking and 0.14 kW for refrigeration.

91
Figure 5.13: Simulated average summer lighting and appliances demand on (a) typical
weekdays and (b) typical weekend

92
 Figure 5.14: Simulated average winter lighting and appliances demand on (a) typical
weekdays and (b) typical weekend

On weekends, the average demand for electric heaters was 0 kW in both seasons, while the
average demand for interior local fans was set at 0.79 kW and 0.19 kW for summer and winter,
respectively. Interior lighting, receptacle equipment, cooking and refrigeration average
93
demands were respectively 1.6 kW, 2.43 kW, 0.09 kW and 0.149 kW in both seasons. The
building’s monthly energy use is presented in Figure 5.15

Figure 5.15: Simulated monthly energy consumption of the building

Based on Figure 5.15, energy consumptions for January, June, July, and December were
relatively low, with an average of 2.48 MWh, which is approximately 50% lower than the
average consumption for the other months. The low consumption is because of holidays, as
outlined in Table 5-5. Heating energy consumption, as indicated by “other process”, was only
noticeable in May, July, August, and September, with an average of 0.50 MWh, which is 94%
lower than indoor local fan average energy consumption. These findings are in line with the
average seasonal outdoor air temperature for Alice (Overen et al., 2019). In general,
receptacle equipment had the highest energy consumption, with a total of 23.80 MWh/year.
This is followed by lighting energy consumption, with a total of 14.64 MWh/year. The
percentage energy-use of appliances and lights is presented in Figure 5.16.

94
 Figure 5.16: Simulated building electrical appliances and lights energy usage and overall
energy use intensity

Receptacle equipment accounts for approximately half of the building’s total energy
consumption, while lighting is responsible for 30%. Interior local fan consumes 17% of the
total energy, and heating is responsible for only 1%. The annual scheduled holidays
significantly reduce the heating energy consumption as the major heating season (in June and
July) was omitted in the simulation. Lastly, cooking and refrigeration account for 2% of total
energy consumption. In general, the overall EUI with respect to conditioned spaces is 31
kWh/m2. As discussed in Section 5.3, non-conditioned spaces were excluded because they
are considered unhabitable spaces and thus do not technically contribute to the building EUI
according to SANS 204 (South African Bureau of Standards, 2011). The annual energy profile
for the building was extracted from Figure 5.15 and used to derive operational energy cost, as
presented in Figure 5.17.

Following the tariff-setting approach discussed in Section 4.2.3, an average energy tariff of
1.12 R/kWh in summer and 2.35 R/kWh in winter was used to derive the building energy cost.
The average tariffs were obtained from the seasonal TOU energy charges in Table 4-5. In
Figure 4.27, the highest energy cost was expected in the winter months given the high demand
charges, but due to holidays, June’s energy cost was relatively low, with R 5 103.26, while the
highest energy cost of R 11 714.52 was observed in August. Similarly, January and December
had the lowest energy cost of R 2 951.87 and R 2 612.18, respectively. This is attributed to
the end-of-year holiday considered in the annual schedule.

95
 Figure 5.17: Simulated monthly energy cost of the building

Moreover, the cumulative annual energy cost for the building is R67 107.75, which is
equivalent to an average cost of R5 592.31/month. The respective annual energy cost of the
various appliances and lights are presented in Figure 5.18.

Figure 5.18: Simulated total energy cost for appliances and lights

96
As shown in Figure 5.18, receptacle equipment and interior lighting energy cost were more
than 50% higher than the energy costs of other appliances. Interior lighting accrued R20
091.86 per year. Seasonally, lighting energy amounted to R54.51 per day in summer and
116.46 per day in winter. However, receptacle equipment had the highest energy cost of R32
702.14 per year, averaging R86.82 per day in summer and R181.87 per day in winter. The
overall daily average operational energy cost was R181.25 in summer and R377.80 in winter.

The resultant GHG emissions impact due to the whole-building energy use is illustrated with
the CO2 emissions bar chart in Figure 5.19. CO2 emissions due to energy-use of appliances
were further consolidated into total system emissions, which represent the interior fan CO2
emissions, other processes (heating), while receptacle equipment CO2 emissions were
captured as total equipment and total lights represent interior lighting CO2 emission.

Figure 5.19: Simulated environmental impact of the building energy consumption

CO2 emission for total lights was 7 596 kg per annum, but total equipment had the highest
emissions of 13 942 kg per annum. System CO2 emissions were the least at 4 201 kg per
annum. Cumulatively, the annual energy consumption of the building was equivalent to 25 440
kg of CO2 emission, which averages at 2 120 kg of CO2 emissions per month.

97
5.4 Lighting Energy Optimisation of the Building: Daylight-linked Interventions

A non-invasive daylight intervention was prioritised for the study in order to avoid interference
with the façade of the building, given its heritage status. In line with Atif and Galasiu (2003),
Alrubaih et al. (2013), and Lim et al. (2017), daylight-linked sensors were adopted in the
research. Automated on/off and continuous dimming daylight interventions were employed to
optimise the building lighting energy consumption. Figure 5.20 shows the automated on/off
and continuous dimming daylight-linked equation graphs used in the simulation.

From Figure 5.20, the automated on/off sensors were programmed to switch off the lights in a
given space when the indoor illuminance is greater than 300 lux and on when the illuminance
is below 300 lux. In other words, the electric lights in the space operate at 0% of their rated
capacity when the indoor illuminance is above 300 lux and at 100% capacity when the
illuminance is below 300 lux. On the other hand, a regression algorithm was adopted in

Figure 5.20: Automatic and continuous dimming equation graphs

continuous dimming sensors. The lights’ operating capacity in a given space decreases from
100% to 0% as the indoor illuminance of the space increase from 0 lux to 300 lux. Further, the
automated on/off and continuous dimming daylight-linked energy performances were
simulated consecutively and compared with the building “as-is” energy performance as
covered in Chapter 4. It is noteworthy that the daylight-linked sensors schedule operates in
parallel with the building’s “as-is” lighting schedule, i.e., the lights must be switched on by the
occupant for the daylight-linked sensors to operate (Lim et al., 2017). Also, in line with Lim et
al. (2017) approach, the sensors were installed on the ceiling facing downward. The graphical
comparison of automated on/off, continuous dimming and “as-is” lighting demand of the

98
building on summer and winter weekdays and weekends is presented in Figures 5.21 and
5.22.

Figure 5.21: Simulated “as-is” and daylight interventions electric lighting demand on (a)
summer weekdays and (b) summer weekends

99
 Figure 5.22: Simulated “as-is” and daylight interventions electric lighting demand on (a)
winter weekdays, and (b) winter weekends

In Figures 5.21 and 5.22, PHY_cond_on-off represents automated on/off lighting demand,
PHY_cond_Dim was used to indicate continuous dimming lighting demand, while PHY_cond
was used to represent the “as-is” lighting demand of the building. On summer weekdays, the
average automated on/off and continuous dimming lighting demands were 1.33 kW and 0.86

100
kW, respectively. This is equivalent to a 47% reduction in lighting demand due to automated
on/off sensors and 65% due to continuous dimming sensors. In summer weekends, the
lighting demand was further reduced to 54% by automatic on/off sensors and 68% by
continuous dimming sensors. In Figures 5.13, 5.14, 5.21 and 5.22, the “as-is” average
weekdays lighting demand and average weekends lighting demand were respectively
constant regardless of the season. This was not the case for automated on/off and continuous
dimming sensors, as their lighting demands vary from one day to another depending on
available solar radiation. Therefore, summer lighting energy demand reduction is expected to
differ from winter demand reduction.

The average winter weekday automated on/off lighting demand was 1.84 kW, equivalent to
26% demand reduction, while continuous dimming lighting demand was 1.50 kW, equivalent
to 40% demand reduction. In winter weekends, automated on/off and continuous dimming
average lighting demand were 0.62 kW and 0.49 kW, respectively, equivalent to 47% demand
reduction for automated on/off sensors and 58% for continuous dimming sensors. A synopsis
of the building “as-is” and the daylight interventions' seasonal and annual lighting demands
are presented in Table 5-6.

Table 5-6: Seasonal and annual “as-is” and daylight interventions lighting energy demand of
the building
Summer Winter Annual
Lights operation Mean Reduction Mean Reduction Mean Reduction
(kW) (%) (kW) (%) (kW) (%)
Automated on/off 0.80 55 0.81 40 0.80 52
Continuous 0.60 66 0.63 53 0.61 64
dimming
No sensor (as-is) 1.77 0 1.36 0 1.67 0

Table 5-6 shows a 15% difference between summer and winter for automated on/off lighting
demand reduction and 13% for continuous dimming lighting energy demand reduction.
Continuous dimming operation had the most lighting demand reduction, with 66% in summer
and 53% in winter. On average, the building’s annual lighting demand was reduced by 52%
with automated on/off sensors and 64% with continuous dimming sensors. The building’s
monthly optimised lighting energy is presented in Figure 5.23.

101
 1.6
1.5
1.4
1.3
1.2
1.1
1
0.9
Energy (MWh)

0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
PHY_cond_on-off.aps PHY_cond_Dim.aps PHY_cond.aps
Figure 5.23: The building’s monthly “as-is” versus optimised light energy consumption

Continuous dimming sensors were found to contribute the least energy consumption of 0.44
MWh per month, while automated on/off sensors contributed 0.59 MWh per month. The
contribution to the building total energy reduction of automated on/off and continuous dimming
was 16% and 20%, respectively, compared to “as-is” lighting energy consumption.

Further, the percentage contribution to the reduction in energy consumption of the automated
and continuous dimming sensors relative to other appliances in the building is presented in
Figure 5.24, which indicates that interior lighting contribution to the building EUI was 17% and
13% due to automated on/off and continuous dimming sensors, respectively.

102
 Interior Local Fans Interior lighting
7.82, 19% 7.02, 17%

Other process
0.61, 1%
Refrigeration
1.18, 3%
Cooking
MWh Automatic on/off
0.18, 2%
EUI:26 kWh/m2

Receptacle Equipment
23.80, 58%

Interior lighting
Interior Local Fans 7.02, 13%
7.62, 19%

Other process
0.65, 2%
Refrigeration
1.18, 3%
MWh
Cooking EUI:25 kWh/m2 Continuous dimming
0.18, 2%

Receptacle Equipment
23.80, 60%

Figure 5.24: The building overall energy-use intensity due to automatic on/off and continuous
dimming daylight-linked control sensors

The use of automated on/off sensors reduced the interior lighting annual cumulative energy
from 14.64 MWh to 7.02 MWh, while continuous dimming sensors reduced overall
consumption to 5.31 MWh. The overall EUI of the building was reduced to 26 kWh/m2 using
automated on/off sensors and was further reduced by 1 kWh/m2 using continuous dimming
sensors.

In terms of the energy costs and benefits, the same tariff structure and procedure used to
deduce the building’s “as-is” energy cost in Chapter 4 was employed to compute the building
optimised lighting energy cost. The building’s monthly “as-is” and optimised energy costs are
presented in Figure 5.25.

In Figure 5.25, the optimised lighting energy cost was 49% less than the “as-is” lighting energy
cost due to automated on/off sensors and 60% less due to continuous dimming sensors. The

103
average summer lighting energy cost due to automated on/off sensors was R650.81 per month
and R486.00 per month due to continuous dimming sensors. Hence, the optimised automated
on/off lighting energy cost in summer is 55% less than the “as-is” lighting energy cost, while
continuous dimming lighting energy cost is 60% less.

Figure 5.25: Monthly “as-is” and optimised lighting energy cost of the building

However, a higher optimised lighting energy cost was observed using both sensors in winter,
implying lower energy cost savings. The winter low energy cost savings is due to the fewer
winter months and the holidays that were factored in the annual building schedules (refer to
Table 5-5). Moreover, in winter, automated on/off and continuous dimming lighting average
energy costs were R1 403.58 and R1 096.04, thus equivalent to 40% and 53% energy cost
savings, respectively.

In addition to the building optimised energy cost benefits evaluation, environmental mitigation
was also assessed. Figure 5.26 shows the mitigated CO2 emissions due to the optimised
lighting energy alongside the “as-is” CO2 emissions.

104
 850
800
750
700
650
600
550
500
450
kgCO2

400
350
300
250
200
150
100
50
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
PHY_cond_on-off.aps PHY_cond_Dim.aps PHY_cond.aps
Figure 5.26: The building optimised lighting energy environmental mitigation

Figure 5.26 shows that automated on/off and continuous dimming sensors resulted in 52%
and 64% CO2 mitigation, respectively. The automated on/off lighting energy consumption was
equivalent to 3 644 kg per annum CO2 emissions, while continuous dimming lighting energy
consumption was equivalent to 2 727 kg per annum CO2 emission. Overall, automated on/off
intervention reduced the whole-building CO2 emission by 16% and 19% with continuous
dimming interventions.

5.5 Conclusions

The simulation-based analyses of the building revealed the impact of the adjacent structure
on the distribution of daylight for indoor spaces. The impact was observed on the luminance
and illuminance of the affected east and west-facing spaces. The “as-is” energy performance
simulation for the entire building reveals that receptacle equipment accounts for the highest
energy consumption at 49%, while electric light was the second-highest energy consumer at
30%. The simulated energy-use intensity of the building was 31 kWh/m2. Receptacle
equipment and lighting exhibit a similar trend in terms of simulated energy cost, with receptacle
equipment accruing R32 702.14 per year while lighting energy amounted to R20 091.86 per
year. The overall building energy cost was R67 108.50 per annum, which averages to
105
R5 592.31 per month and R43.00/m2. CO2 emission for total lights was 7 596 kg per annum,
but total equipment had the highest emissions at 13 942 kg per annum. System CO2 emission
was the lowest at 4 201 kg per annum. The overall annual emission for the building is 25 440
kg which averages to 2 120 kg per month.

Data and analyses indicate that the use of automated on/off sensors in the building could
reduce lighting energy demand for summer weekdays by 1.33 kW and winter weekdays
lighting demand by 1.84 kW. This demand reduction is equivalent to 47% and 26% on summer
and winter weekdays, respectively. The reduction in lighting energy demand resulted in 52%
of lighting energy savings and 16% of whole-building energy and energy-use intensity (EUI)
reductions. The energy cost benefits of automated on/off intervention were found to be 50%
per annum, i.e. the “as-is” lighting energy cost was reduced from R20 091.89 per annum to
R10 068.11 per annum. The environmental impact of the building lighting and whole-building
energy consumption was mitigated by 52% and 16%, respectively.

In terms of the use of continuous dimming sensors as daylighting intervention, summer and
winter weekdays' average lighting demand was reduced by 0.86 kW and 1.50 kW,
respectively. This is equivalent to a 65% lighting demand reduction in summer and 58% in
winter. The resultant lighting annual energy savings is 9.32 MWh, which is equal to 64%, while
the whole-building demand and EUI were reduced by approximately 20%. Continuous
dimming sensors were found to reduce the lighting energy cost by 62%, thus saving
R12 420.73 per year. Mitigation of CO2 emissions due to the lighting annual energy usage and
the whole-building energy consumption were reduced by 64% and 19%, respectively.

The findings systematically demonstrate that continuous dimming daylight-linked is the most
feasible daylight intervention. The findings also show that daylighting can be used to optimise
the building's energy performance. Most importantly, the findings strongly demonstrate that
the lights should be controlled with daylight-linked sensors in order to achieve the best
optimisation in energy performance. However, this also requires the effective operation of
window covers through the choice of appropriate blinds and curtains.

106
 6CHAPTER 6

Consolidation of Sub-findings, Conclusions and Recommendations

6.1 Introduction

This study evaluates options in daylighting interventions for a heritage instruction-cum-office

building with the objective of reducing overall energy-use intensity (EUI), related operational
energy cost and mitigation of CO2 emissions. The study demonstrates the potential of heritage
buildings to utilise daylight for energy efficiency without altering the original structure. The
purpose of the study was also to show how buildings completed before 1999 can be
considered for energy efficiency initiatives in South Africa in contrast to Energy Performance
Certificate (EPC) regulations that currently exclude such buildings from the EPC initiative for
the time being.

The study hypothesised that heritage buildings could utilise daylighting to save up to 50% on
lighting energy usage and resultant operational cost and mitigate CO2 emissions using non-
invasive daylight interventions. Guided by four sub-questions, the study derived related sub-
findings which are consolidated in this Chapter in order to derive overall findings on the main
research question. The first sub-question focused on the status quo (as-is) EUI for the selected
case study building, while the second one focused on key daylighting interventions for the
case study building. The third sub-question addressed the “as-is” contribution of daylighting to
the EUI of the building, and the final one focused on feasible retrofits for daylighting
interventions toward reducing EUI as well as the related viability of such interventions. Sub-
findings on the sub-questions were derived from analyses of primary and secondary data.
Primary data were mainly generated by experimental and simulation assessments of the
building’s daylighting and energy performance. This Chapter covers the consolidation of the
sub-findings into the overall findings of the study, as well as overall conclusions and
recommendations for further studies.

6.2 The “as-is” Energy-use Intensity of the Building

Sub-question one was addressed through whole-building energy performance simulation as

covered in Chapter 4. Secondary data on floor area, the building thermal component
dimensions, and optical properties were used to develop the building model. The primary data
on the number of lights and appliances in the various spaces of the building were also
collected. Secondary data consists of the nameplate power rating of the lights and appliances
as well as the occupancy period for the spaces. Occupancy periods of spaces relative to the

107
dominant uses of the building were obtained from the 2018 prospectus of the Faculty of
Science and Agriculture at the University of Fort Hare. This was to ensure that the normal
academic schedule of the building was used as opposed to its use schedule during the Covid-
19 pandemic period. Secondary and primary data were used to schedule annual operations
for lighting and appliances as one of the key inputs for the “as-is” and optimised lighting energy
performance simulation for the building.

From the simulation, the whole-building annual energy consumption was derived to be
49.02 MWh. Receptacle equipment was the highest energy consumer at 23.8 MWh per
annum, while lighting had the second-highest consumption at 14.64 MWh per annum. “As-is”
EUI was derived to be 31 kWh/m2/yr at a resultant operational energy cost of R67 108.50 per
annum. In terms of environmental impact, the building’s annual energy consumption is
equivalent to 25 440 kg CO2 emissions.

6.3 The Building Key Daylighting Interventions

From the walk-through audit, no specific daylighting interventions for the building were found.
The audit established that the building has a conventional design with external side-lighting
(vertical windows) that admit natural lights and ventilation at the same time. However, the
majority of the windows have vertical fabric blinds, while one of the spaces had venetian
aluminium blinds. Spaces with vertical fabric blinds were found to have better daylight
illuminance than the space with venetian aluminium blinds. Nevertheless, the building daylight
simulation (which excluded window covers) revealed that the indoor average daylight
illuminance of the ground floor spaces ranges between a high of 6 347.02 lux and a low of
23.01 lux, while the top floor spaces' average illuminance ranges from a low of 38.37 lux to a
high of 6 411.59 lux.

Building orientation relative to the daily sun trajectory and adjacent structure is the most critical
factor influencing the irregular indoor daylight distribution of the building. Based on the
findings, and with secondary data from Atif and Galasiu (2003), Lim et al. (2017), and Yu et
al. (2015), automated on/off and continuous dimming daylight-linked sensors were identified
and used as daylighting-related interventions in the building. Automated on/off daylight-linked
sensor controls electric lights based on a pre-set ambient light level. The sensor automatically
switches off electric lights when the pre-set ambient light level is met and switches on the
lights when the indoor illuminance is below the pre-set ambient light level. Continuous dimming
daylight-linked sensor also controls electric lights using ambient light level by gradually

108
decreasing the power demand of electric lights as the indoor illuminance increases and vice
versa.

6.4 “As-is” Daylighting Contribution to the Building Energy-use Intensity

Sub-question 3 deals with the energy savings due to the building’s “as-is” daylighting. As
indicated in Section 6.2, no daylighting retrofit interventions were evident in the building during
the walk-through audit. However, the experimental aspect of the research targeted the
potential lighting energy savings if daylight-linked sensors were installed in the building to
control electric lights. Unlike the simulation evaluation of the lighting performance of the
building, the experimental evaluation included the effect of the window covers such that the
measured daylight illuminance is the “as-is” operating conditions for the respective spaces
where measurements were taken (due to the limited number of sensors only six spaces out of
33 conditioned spaces and 68 including unconditioned spaces were sampled for the
experiment-based component of the study).

The experiment-based component also entailed an onset HOBO photometric sensor placed
on the desk in each sampled space to monitor the daylight illuminance sensed by an occupant
sitting at the desk. The experiment was conducted from April to July, with April and May
representing summer while June and July represented winter. Experimental data revealed that
the average daylight illuminance of three of the spaces sampled was below the South African
workplace light threshold of 300 lux, while the average illuminance of the other three spaces
was above the 300 lux threshold. This implies that only the spaces with average daylight
illuminance above 300 lux can utilise daylight-linked sensors for lighting energy efficiency. In
the winter season, spaces with daylight illuminance above 300 lux were found to have the
potential of 1.61 kWh lighting energy savings on weekdays, approximately 15% savings. On
weekends, the lighting energy savings were up to 52%. Summer lighting energy was observed
to be the least, with a maximum savings of 33% on weekdays and 24% on weekends. The
resultant lighting energy cost savings for the six spaces were 57% in summer and 25% in
winter.

6.5 Feasible Daylighting Interventions for Energy-use Reduction

As indicated in Section 6.2, automated on/off and continuous dimming daylight interventions
were simulated towards optimisation of lighting energy performance and resultant EUI of the
building. Sub-question 4 was addressed using simulation, and thus the primary and secondary
data in section 6.1 were used in developing the building model. For purposes of the simulation,
both sensor-types were modelled facing downward from the ceiling and executed sequentially.

109
Continuous dimming daylight intervention resulted in more energy savings than automated
on/off daylight intervention. Continuous dimming sensors reduced the annual energy-use for
lighting by 64% and overall annual energy-use by 20%, while overall EUI was reduced by
19%. The resultant overall energy cost saving was R12 420.73 per annum, while the mitigated
CO2 emission was 4 839 kg per annum.

6.6 Consolidation of Findings and Conclusions

From the findings, the study hypothesis that heritage buildings can utilise daylighting for up to
50% lighting energy savings and resultant operational cost and CO2 emissions is provisionally
validated, bearing in mind the simulated results of the study. As illustrated in Chapter 4, three
out of six spaces sampled have sufficient daylight illuminance for typical office tasks based on
the South African workplace light level recommendation. The spaces were found to have the
potential to optimise daylight utilisation to save up to 33% lighting energy in summer and 52%
in winter. The resultant energy cost-benefit was up to 57% in summer and 25% in winter. A
graphic summary of the experimental findings of the research is presented in Figures 6.1 and
6.2.

Figure 6.1: Daylighting performance summary of selected spaces on the building’s ground
floor

However, the whole-building daylighting simulation revealed that the maximum illuminance of
the building is 16 411.54 lux. The relatively high indoor daylight illuminance is due to the

110
exclusion of the window covers in the simulation. This implies that a significant amount of
daylight can be harvested in the building with an effective operation of the window covers and
choice of window cover materials. Opening the windows during the day or using dynamic
photosensors to operate the windows would result in significant improvements in daylight
illuminance indoors.

Figure 6.2: Daylighting performance summary of selected spaces on the building's top floor

Moreover, simulation for optimised lighting energy performance, as reported in Chapter 5,

shows that using automated on/off sensors reduced the annual lighting energy by 7.61 MWh
(52% energy savings) and 9.32 MWh (64% energy savings) due to continuous dimming.
Similarly, automated on/off sensors reduced the whole-building annual energy consumption
and EUI by 7.77 MWh and 5.0 kWh/m2/yr, respectively, which is equivalent to a 16% reduction
in each case. In terms of continuous dimming, the whole-building annual energy was reduced
by 9.4 MWh (20% savings), while the EUI was reduced by 6.0 kWh/m2/yr (19% reduction).
Lighting and energy performance simulation further ascertain the potential of utilising daylight
for lighting and whole-building energy savings. Lighting interventions also reduced operational
energy costs and mitigated CO2 emissions. The use of automated on/off sensors reduced the
lighting energy cost by 50%, and a 62% reduction was observed for continuous dimming
sensors. Further, CO2 emissions due to annual lighting energy consumption were mitigated
by 52% and 64% using automated on/off and continuous dimming sensors, respectively.

111
Automated on/off sensors reduced annual CO2 emissions for the entire building by 16% and
19% for continuous dimming sensors.

6.7 Limitations and Recommendations

The experimental evaluation of the building, which captured the building’s daylight
performance at operating conditions, was limited to six spaces (offices) due to limited
photometric sensors. The experimental observation was also limited to the daylight falling on
the desk in the respective spaces. The spot measurement adopted in the experimental
approach was insufficient to quantify the distribution and the amount of daylight in the entire
space compared to the experimental and simulation findings. Although the simulation
approach covered the daylighting distribution in a given space and the entire building, it was
limited to design conditions. In other words, human interference, such as the choice of window
covers, operation of windows covers and electric lights, were not considered. Also, the optical
properties of the windows in the model were based on the existing database in the IES
application, which may vary from the building’s actual windows. Further, the experimental and
simulation findings are limited to Alice’s weather conditions since the research was conducted
in Alice only. Occupants’ perception of the indoor daylight level is another crucial aspect
regarding the use of daylight for lighting energy efficiency that was not considered in both
approaches of the research.

Regarding the building energy performance evaluation, the lights’ operation in the
experimental approach was based on measured illuminance data, while the simulated light
operation was based on classes schedule. In addition, the power factor and lamp decay factor
were not considered in the lighting energy computation. The computed lighting energy was
based on active power.

In view of the outlined limitations, the use of multiple photometric sensors for the experimental
evaluation of daylight is recommended. As alluded to by Kruisselbrink, Dangol and Rosemann
(2018), photometric sensors in daylight evaluation should be set up in a grid that covers at
least the area of interest in the space. Figure 6.3 shows a schematic diagram of the daylight
evaluation experimental approach.

112
Figure 6.3: Experimental daylight evaluation approach

However, this approach increases the cost and time of the research, thus, the simulation
approach is a cost and time mitigating method for whole-building daylight experimental
evaluation (Yu et al., 2015). Moreover, it is recommended that window covers be integrated
into building modelling to simulate daylighting performance with the effect of blinds to
demonstrate daylight performance under operating conditions. The feasibility of adopting
daylight for lighting energy savings in other heritage buildings in South Africa is a study that
should be considered in the future. Such studies would create a benchmark for energy
performances and lighting energy efficiency regulation of heritage buildings in the country.
Further, occupants’ perception of daylight level needs to be considered in future studies as it
can be used to reduce workplace light requirement threshold, which will result in an increase
in the lighting energy saving and overall energy EUI. Arising from the delineation (Chapter 1,
Section 1.10), further study on allowable fenestration changes and related simulation for
daylight performance should be undertaken to assess the extent to which fixed fenestration
influences indoor daylight optimisation.

Furthermore, by comparison of experimental and simulation findings, the use of continuous

dimming sensors and effective operation of window covers would result in energy savings in
commercial buildings regardless of age. Modern designs incorporate dynamic photosensors
in buildings to aid the operation of window covers based on outdoor and indoor illuminances.
The use of non-invasive dynamic (photo-automated) window covers in heritage buildings
would therefore result in enhanced indoor daylighting and, at the same time, reduce glare and
visual discomfort by preventing an excessive amount of sunlight penetration. In addition, as
observed in Chapter 4, window cover materials with low transmissivities, such as aluminium

113
blinds, are likely to result in poor indoor daylighting levels as well as low uniformity in
distribution.

However, the choice of window cover materials should be influenced by the dominant task
lighting requirement for the space. Further, a side-lighting design strategy where the same
windows are used for daylighting and ventilation is one of the setbacks of daylighting in
heritage buildings. As indicated in Chapter 4, side-lighting is commonly associated with poor
indoor daylight quality due to low uniformity and diversity, which in turn results in visual
discomfort for many occupants. Building orientation, adjacent structures, and indoor furniture
are the main influencing factors of the quality of light in side-lighting design strategies.
Therefore, effective daylighting in heritage buildings with side-lighting is a factor of the distance
of the area of interest from the external windows as shown in the building’s annual sunlight
exposure in Figure 6.4.

Figure 6.4: Annual sunlight exposure of the building; (a) top floor and (b) ground floor

114
A way of addressing this challenge without altering the structure would be through
repurposing floor areas by placing tasks with high light level requirements in spaces with larger
window areas, closer to the building perimeter, and east, north or west facing depending on
the dominant task period. The occupants sitting or working position with respect to the
direction of direct sun rays should be taken into consideration to avoid glare.

115
 References

Alrubaih, M.S., Zain, M.F.M., Alghoul, M.A., Ibrahim, N.L.N., Shameri, M.A. and Elayeb, O.
2013. Research and development on aspects of daylighting fundamentals. Renewable and
Sustainable Energy Reviews, 21, 494–505.
Akimov, L., De Michele, G., Filippi Oberegger, U., Badenko, V. and Mainini, A.G. 2021.
Evaluation of EN15193-1 on energy requirements for artificial lighting against Radiance-based
DAYSIM. Journal of Building Engineering. 40(January):102698.
Al-Obaidi, K.M., Ismail, M.A. and Abdul Rahman, A.M. 2015. Assessing the allowable daylight
illuminance from skylights in single-storey buildings in Malaysia: a review. International
Journal of Sustainable Building Technology and Urban Development. 6(4):236–248.
Alrubaih, M.S., Zain, M.F.M., Alghoul, M.A., Ibrahim, N.L.N., Shameri, M.A. and Elayeb, O.
2013. Research and development on aspects of daylighting fundamentals. Renewable and
Sustainable Energy Reviews. 21:494–505.
Asif Ul Haq, M., Islam, A., Shihavuddin, A.S.M., Maruf, M.H., Al Mansur, A. and Hassan, M.Y.
2020. Enhanced energy savings in indoor environments with effective daylight utilisation and
area segregation. Symmetry. 12(8):1–19.
Atif, M.R. and Galasiu, A.D. 2003. Energy performance of daylight-linked automatic lighting
control systems in large atrium spaces: Report on two field-monitored case studies. Energy
and Buildings. 35(5):441–461.
Ash, P., 2021. Be ready for another five years of load-shedding, Eskom warns. Times Live, 1
March.
de Bakker, C., Aarts, M., Kort, H. and Rosemann, A. 2018. The feasibility of highly granular
lighting control in open-plan offices: Exploring the comfort and energy saving potential.
Building and Environment. 142(09):427–438.
Bernett, A., Kral, K. and Dogan, T. 2021. Sustainability evaluation for early design (SEED)
framework for energy use, embodied carbon, cost, and daylighting assessment. Journal of
Building Performance Simulation, 14(2), 95–115.
Bian, Y. and Ma, Y. 2017. Analysis of daylight metrics of side-lit room in Canton, South China:
A comparison between daylight autonomy and daylight factor. Energy and Buildings, 138,
347–354.
Boodu, V. 2021. Zone of Space for Natural Lighting. in South African Institute of Building
Design Vol. 04. Durban: South African Institute of Building Design. 1–53.
Booysen, M.J., Samuels, J.A. and Grobbelaar, S.S. 2021. LED there be light: The impact of
replacing lights at schools in South Africa. Energy and Buildings. 235(1):1–11.
Bradshaw, V. 2006. The Building Environment: Active and Passive Control Systems. John
Wiley & Sons.
Chang, W.K. and Hong, T. 2013. Statistical analysis and modeling of occupancy patterns in
open-plan offices using measured lighting-switch data. Building Simulation. 6(1):23–32.
Chi, D.A., Moreno, D. and Navarro, J. 2018. Correlating daylight availability metric with
lighting, heating and cooling energy consumptions. Building and Environment. 132(September
2017):170–180.
Chinazzo, G., Wienold, J. and Andersen, M. 2019. Daylight affects human thermal perception.
Scientific Reports, 9(1), 1–15.

116
Choi, J.-H., Loftness, V. and Aziz, A. 2012. Post-occupancy evaluation of 20 office buildings
as basis for future IEQ standards and guidelines. Energy and buildings, 46, 167–175.
Department of Labour. 1993. Environmental Regulations for Workplaces, Department of
Labour, Pretoria.
Department of Mineral Resources and Energy, 2020. Regulations for the mandatory display
and submission of energy performance certificates for buildings. Ministry of Mineral Resources
and Energy, Pretoria.
Doulos, L., Tsangrassoulis, A. and Topalis, F. 2008. Quantifying energy savings in daylight
responsive systems: The role of dimming electronic ballasts. Energy and Buildings. 40(1):36–
50.
Elakkad, N. and Ismaeel, W.S.E. 2021. Coupling performance-prescriptive based daylighting
principles for office buildings: Case study from Egypt. Ain Shams Engineering Journal. 1–11.
Eskom, 2013. The Eskom National Efficient Lighting Programme: Compact Fluorescent
Lamps (CFL) Clean Development Mechanism (CDM) Project. Eskom, Johannesburg.
Eskom, 2015. Lighting and sensors: Reducing energy costs in commercial buildings. Eskom,
Johannesburg.
Eskom, 2019. Integrated report. Eskom, Johannesburg.
Eskom, 2020. Integrated Report 2021. Eskom, Johannesburg.
Eskom, 2021. Tariffs & charges booklet 2021/22, Eskom, Johannesburg.
Etzinger, A. 2014. Systems engineering approach to Integrated Demand-side Management -
An Eskom case study. Electricity+Control, 09(09), 42–45.
European Committee for Standardisation. 2017. Energy performance of buildings - Energy
requirements for lighting - Part 1: Specifications, Module M9. European Union.
Fawkes, H. 2005. 3. Incentives for energy efficiency in South African industry. Journal of
Southern Africa, 16(4), 18–25.
Freewan, A.A.Y. and Al Dalala, J.A. 2020. Assessment of daylight performance of Advanced
Daylighting Strategies in Large University Classrooms; Case Study Classrooms at JUST.
Alexandria Engineering Journal. 59(2):791–802.
Gentile, N., Dubois, M.C., Osterhaus, W., Stoffer, S., Amorim, C.N.D., Geisler-Moroder, D.
and Jakobiak, R. 2016. A toolbox to evaluate non-residential lighting and daylighting retrofit in
practice. Energy and Buildings. 123:151–161.
Hinrichs, R. and Kleinbach, M. 2013. Energy: Its use and the environment. BROOKS/COLE,
New York.
Hong Kong Green Building Council Limited. 2021. Green Building Design Building Design
Best Practice Guidebook. Hong Kong.
IEA and UNEP. 2019. 2019 Global Status Report for Buildings and Construction: Towards a
zero-emission, efficient and resilient buildings and construction sector, International Energy
Agency, Paris.
International Energy Agency. 2021. Net Zero by 2050: A Roadmap for the Global Energy
Sector. Paris.
Kota, S., Haberl, J.S., Clayton, M.J. and Yan, W. 2014. Building Information Modeling (BIM)-
based daylighting simulation and analysis. Energy and Buildings. 81:391–403.

117
Krarti, M., Erickson, P.M. and Hillman, T.C. 2005. A simplified method to estimate energy
savings of artificial lighting use from daylighting. Building and Environment. 40(6):747–754.
Kreider, J.F., Curtiss, P.S. and Rabl, A. 2009. Heating and cooling of buildings: design for
efficiency. CRC Press, Denver.
Krüger, E.L. and Dorigo, A.L. 2008. Daylighting analysis in a public school in Curitiba, Brazil.
Renewable Energy, 33(7), 1695–1702.
Kruisselbrink, T., Dangol, R. and Rosemann, A. 2018. Photometric measurements of lighting
quality: An overview. Building and Environment. 138(02):42–52.
Kunwar, N., Cetin, K.S., Passe, U., Zhou, X. and Li, Y. 2020. Energy savings and daylighting
evaluation of dynamic venetian blinds and lighting through full-scale experimental testing.
Energy, 197(2), 1–13.
Kwong, Q.J. 2020. Light level, visual comfort and lighting energy savings potential in a green-
certified high-rise building. Journal of Building Engineering, 29(10), 1–9.
Lee, P., Lam, P.T.I. and Lee, W.L. 2018. Performance risks of lighting retrofit in Energy
Performance Contracting projects. Energy for Sustainable Development. 45:219–229.
Li, H., Wu, D., Yuan, Y. and Zuo, L. 2021. Evaluation methods of the daylight performance
and potential energy saving of tubular daylight guide systems: A review. Indoor and Built
Environment, 0(111), 1–17.
Lim, G.H., Keumala, N. and Ghafar, N.A. 2017. Energy saving potential and visual comfort of
task light usage for offices in Malaysia. Energy and Buildings, 147, 166–175.
Mantzourani, K., Doulos, L.T., Kontadakis, A. and Tsangrassoulis, A. 2020. The effect of the
daylight zone on lighting energy savings. IOP Conference Series: Earth and Environmental
Science, 410(1), 1–11.
Michael, A., Gregoriou, S. and Kalogirou, S.A. 2018. Environmental assessment of an
integrated adaptive system for the improvement of indoor visual comfort of existing buildings.
Renewable Energy, 115, 620–633.
Milford, R. 2015. SAN 1544 Energy Performance Certificates - A Driver of Energy Efficiency.
Pretoria.
Onaygil, S. and Güler, Ö. 2003. Determination of the energy saving by daylight responsive
lighting control systems with an example from Istanbul. Building and Environment, 38(7), 973–
977.
Öner, M., Kazanasmaz, T., Leccese, F. and Salvadori, G. 2020. Analysis of the relationship
between daylight illuminance and cognitive, affective and physiological changes in visual
display terminal workers. Building Services Engineering Research and Technology, 41(2),
167–182.
Onset. 2018. HOBO U12-012 Data Logger. Bourne.
Overen, O.K., Meyer, E.L. and Makaka, G. 2021a. Passive Solar and Conventional Housing
Design: A Comparative Study of Daylighting Energy Efficiency Potential. Frontiers in Energy
Research, 9(5), 1–11.
Overen, O.K., Meyer, E.L. and Makaka, G. 2021b. Indoor Daylighting and Thermal Response
of a Passive Solar Building to Selective Components of Solar Radiation. Buildings, 11(1), 1–
18.

118
Overen, O.K. and Meyer, E.L. 2019. Daylighting: A Residential Energy - efficiency Demand-
side Management Initiative, 27th IEEE Domestic Use of Energy, Cape Peninusula University
of Technology, Wellington. 25–27 March.
Roisin, B., Bodart, M., Deneyer, A. and D’Herdt, P. 2008. Lighting energy savings in offices
using different control systems and their real consumption. Energy and Buildings, 40(4), 514–
523.
Saha, H., Florita, A.R., Henze, G.P. and Sarkar, S. 2019. Occupancy sensing in buildings: A
review of data analytics approaches. Energy and Buildings. 188–189:278–285.
Shailesh, K.R. and Raikar, T.S. 2010. Application of RELUX software in simulation and
analysis of energy efficient lighting scheme. International Journal of Computer Applications,
9(7), 24–35.
SOLARGIS. 2017, Solar resource maps of South Africa: Photovoltaic Electricity Potential,
Internet: www.solargis.com/maps-and-gis-data/download/south-africa, last accessed 25 June
2021.
South African Bureau of Standards, 2011. SANS 204: SOUTH AFRICAN NATIONAL
STANDARD: Energy Efficiency in Buildings. Department of Energy, Pretoria.
South African Bureau of Standards. 2021. South African National Standard. The application
of National building regulations. Part X: Environmental Sustainability. Part XA: Energy usage
in buildings. Pretoria.
South Africa Heritage Resources Agency. 1999. National Heritage Resources Act (Act 25 of
1999). Pretoria
Szokolay, S.V., 2004. Introduction to architectural science: the basis of sustainable design.
Architectural Press, Oxford.
Szokolay, S.V. 2008. Introduction to architectural science: the basis of sustainable design.
Second ed. Vol. 8. C. Brisbin (ed.). Oxford: Elsevier.
Tregenza, P. and Mardaljevic, J. 2018. Daylighting buildings: Standards and the needs of the
designer. Lighting Research and Technology, 50(1), 63–79.
UNFCCC. 2008. EB41 - Annex 11: Methodological Tool: Tool to calculate project or leakage
CO2 emissions from fossil fuel combustion v. 02. [Online], Available:
https://cdm.unfccc.int/EB/index.html.
Wirz-Justice, A., Skene, D.J. and Münch, M. 2020. The relevance of daylight for humans.
Biochemical Pharmacology. 1–4.
World Energy Council. 2020. World Energy: Decoding New Signals of Change, World Energy
Council, London.
Yu, X. and Su, Y. 2015. Daylight availability assessment and its potential energy saving
estimation -A literature review. Renewable and Sustainable Energy Reviews, 52, 494–503.
Zhang, W. and Lu, L. 2019. Overall energy assessment of semi-transparent photovoltaic
insulated glass units for building integration under different climate conditions. Renewable
Energy, 134, 818–827.

119
Appendix A-1: Inventory of ground floor lights and appliances

120
Appendix A-2: Inventory of top floor lights and appliances

121
Appendix B-1: Ground floor spaces solar shading

122
123
124
125
126
Appendix B-2: Top floor spaces solar shading

127
128
129
130
131
Appendix C-1: Ground floor spaces GVCP and DGI

132
Appendix C-2: Top floor spaces GVCP and DGI

133
Appendix D-1: Ethics clearance

134