University of Southern California Dissertations and Theses

Evaluation and development of solar control screens: using daylight simulation to improve the performance of facade solar control screens

(USC Thesis Other)

Evaluation and development of solar control screens: using daylight simulation to improve the performance of facade solar control screens

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content pg. 1

EVALUATION AND DEVELOPMENT OF SOLAR CONTROL SCREENS:
USING DAYLIGHT SIMULATION TO IMPROVE THE PERFORMANCE OF
FACADE SOLAR CONTROL SCREENS

BY
Hui Ling Chang

A Thesis Presented to the
FACULTY OF THE USC SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial Fulfillment of the Requirements for the
MASTER DEGREE OF BUILDING SCIENCE

May 2015

pg. 2
ACKNOWLEDGEMENT
I would like first and foremost to acknowledge and thank Professor Kyle Konis for his entirely
involvement in this project. He guided me to this meaningful research and supported me to learn
the technical skills. He offered great help in shaping the whole concept and organizing the
structures. Professor Konis leaded me to the daylighting area that I was not familiar with, and I
not only learned the knowledge of daylighting but also started to think about the difference of
viewpoints between architects and engineering. This concept would be very helpful to apply in
my career in the future.
I would also like to acknowledge Professor Marc Schiler for his support. Professor Schiler
reminded me the key point to clarify critical concept in my research, and his suggestions always
helped me out whenever there was an obstacle. I really appreciated his involvement to make me
get through the whole process fluently and I also learned how to think those complex questions
in the right way because of his suggestions.
Professor Noble has been paid a lot of effort on grouping the MBS family and encouraging us in
the past two years. As an international student, it is not easy to involve into a new culture;
nevertheless, because of Professor Noble’s effort, I experienced many activities in American
culture and I always got supports from anyone in our MBS family. The every moment here was
amazing.
Besides, I would like to thank Hu Guoyu who always gave me the greatest support whenever I
face technical difficulties. When I struggled with problems, his advice helped me simplify the
problems and solve it efficiently. I would like to thank Shang Sun who assisted me to improve
the structure of my thesis in the final review. For the whole MBS in the second year, Chao, Gary,
Fahad, Geoffrey, Sebanti, Spurthy, Mohammad and Yibo, you are not only my classmates but
also are my good friends. I am so glad to join MBS group and meet all of you that I had so many
enjoyable moments.
Finally, I am very a very lucky person, because I could not have all of the priceless experiences
in USA without my family support. I appreciate they are always with me, and I love you so
much.

pg. 3
Table Contents

ABSTRACT .................................................................................................................................. 11
1 INTRODUCTION ................................................................................................................ 12
1.1 Hypothesis ...................................................................................................................... 13
1.2 Importance ...................................................................................................................... 13
1.3 Present Problem.............................................................................................................. 13
1.4 External conditions ......................................................................................................... 14
1.4.1 Orientation/location ................................................................................................ 14
1.4.2 Sky Condition ......................................................................................................... 15
1.5 Existing Complex Fenestration Design .......................................................................... 15
1.5.1 Novel louvers .......................................................................................................... 15
1.5.2 Perforated screen ..................................................................................................... 16
1.6 Contribution ................................................................................................................... 18
1.7 Study Boundaries ........................................................................................................... 19
1.7.1 Deliverable .............................................................................................................. 20
1.7.2 Outlines of the Study .............................................................................................. 20
2 BACKGROUND RESEARCH ............................................................................................ 21
2.1 Case study of an existing building ................................................................................. 21
2.2 3D Parametric Geometry ................................................................................................ 22
2.3 Daylight Glare Probability ............................................................................................. 24
2.4 Climate-based Daylight Metric, Useful Daylight Illuminance. ..................................... 26
2.5 Climate-based Daylight Metric, Spatial Daylight Autonomy ........................................ 27
2.6 Solar Radiation ............................................................................................................... 29
2.7 Material Properties ......................................................................................................... 31
pg. 4
3 METHODOLOGY ............................................................................................................... 33
3.1 Modeling Tools .............................................................................................................. 34
3.1.1 Ladybug and Honeybee .......................................................................................... 34
3.1.2 Rendering in Radiance ............................................................................................ 36
3.2 Variables......................................................................................................................... 38
3.3 Geometry ........................................................................................................................ 38
3.3.1 Base Case ................................................................................................................ 38
3.3.2 Louvers ................................................................................................................... 39
3.4 Key parameters ............................................................................................................... 40
3.4.1 View ........................................................................................................................ 40
3.4.2 Daylight glare probability ....................................................................................... 41
3.4.3 Useful daylight illuminance .................................................................................... 42
3.4.4 Solar Radiation........................................................................................................ 42
3.5 Comparison of Measurements ........................................................................................ 43
3.6 Performance Evaluation ................................................................................................. 44
4 RESULTS OF MESUREMENTS ........................................................................................ 47
4.1 Opening percentage ........................................................................................................ 47
4.2 Base case ........................................................................................................................ 49
4.3 Louvers ........................................................................................................................... 51
4.3.1 Daylight Glare Probability ...................................................................................... 51
4.3.2 Useful Daylight Illuminance ................................................................................... 57
4.3.3 Solar radiation ......................................................................................................... 61
4.4 Effect of latitude on DGP and UDI ................................................................................ 62
4.4.1 Evaluation of latitude effect on DGP ...................................................................... 62
4.4.2 Evaluation of latitude effect on UDI ....................................................................... 66
pg. 5
4.5 Effect of material effect on DGP and UDI ..................................................................... 66
4.5.1 Evaluation of material effect on DGP ..................................................................... 66
4.5.2 Evaluation of material effect on UDI...................................................................... 67
5 EVALUATION OF OVERALL PERFORMANCE ............................................................ 71
5.1 Ranking of Geometries by linear transformation ........................................................... 71
5.1.1 Ranking of single numerical score .......................................................................... 71
5.1.2 Ranking of single parameter by weighted score ..................................................... 73
5.1.3 Ranking of double parameters by weighted score .................................................. 74
5.2 Performance examination by paired comparison ........................................................... 77
5.2.1 Performance examination based on viewpoint of architecture ............................... 78
5.3 Performance examination based on viewpoint of energy .............................................. 80
6 CONCLUSION ..................................................................................................................... 83
7 FUTURE WORK .................................................................................................................. 85
8 BIBLIOGRAPHY ................................................................................................................. 86
9 APPENDIX ........................................................................................................................... 91
9.1 False color of HDR imaging in base case ...................................................................... 91
9.2 False color of HDR imaging in louvers within variable angles ..................................... 92
List of Figures
Figure 1.1 Glare from perforated screen (the De Young museum by Herzog and Demeuron) .... 14
Figure 1.2 Example and Application of RETRO Lux .................................................................. 16
Figure 1.3 Example of Lightlouver ............................................................................................... 16
Figure 1.4 View through MSPSS (left), unobstructed view (right) (David Appelfeld, Andrew
McNeil and Svend Svendsen 2012) .............................................................................................. 17
Figure 1.5 Effect of expanded metals on solar transmission and eye’s view ............................... 18
Figure 1.6 New Museum of Contemporary Art of New York’s rhombus-shaped expanded metal
mesh .............................................................................................................................................. 18
pg. 6
Figure 1.7 COLT Solarlouver External Solar Shading System at HSE Bootle in UK ................. 19
Figure 2.1 Example of transmission of direct sunlight and personal modification on SE façade 22
Figure 2.2 Compressed screen at 3 PM in four seasons (Nancy Cheng, 2012) ............................ 23
Figure 2.3 Tensioned screen at 3 PM in four seasons (Nancy Cheng, 2012) ............................... 23
Figure 2.4 The rotation of unit configuration and different rotation angles form 0 to 90 degrees
(Y. Elghazi et al., 2014) ................................................................................................................ 28
Figure 2.5 The optimization of Kaleirdocycle configuration in daylight performance (Y. Elghazi
et al., 2014) ................................................................................................................................... 29
Figure 2.6 Temperature evolution inside offices during a winter day with clear sky conditions
and external dry bulb temperature from 0 ◦C to 20 ◦C, for a building with 50% WWR (Alan Pino
et al., 2012) ................................................................................................................................... 30
Figure 2.7 Box plot of total annual energy demand with respect to use of solar protection devices
(Alan Pino et al., 2012) ................................................................................................................. 31
Figure 2.8 Reflection process of specular and diffuse materials .................................................. 32
Figure 2.9 Simulated view of a room with walls made of diffuse oak (specularity 0.0%),
varnished checkmate oak (specularity 0.2%) and varnished shining oak (specularity 2.3%) (M.
Bodart et al, 2007)......................................................................................................................... 32
Figure 3.1 Size of shoebox in modeling ....................................................................................... 34
Figure 3.2 Components in Ladybug ............................................................................................. 35
Figure 3.3 Component in Honeybee ............................................................................................. 35
Figure 3.4 Components for real time examination ....................................................................... 36
Figure 3.5 False color HDR image of plastic louver in 15˚ at 12 PM on winter solstice in Los
Angeles ......................................................................................................................................... 38
Figure 3.6 Geometry of base case ................................................................................................. 39
Figure 3.7 Geometry of louvers .................................................................................................... 40
Figure 3.8 Titling angles downward from 0 ˚ to 75 ˚ ................................................................... 40
Figure 3.9 Way of calculating occlusion ratio .............................................................................. 41
Figure 3.10 Observing point of DGP in shoebox ......................................................................... 41
Figure 3.11 Measured surface for UDI in shoebox ....................................................................... 42
Figure 3.12 Component of Radiation Analysis............................................................................. 43
Figure 3.13 Vertical surface measuring solar radiation ................................................................ 43
pg. 7
Figure 3.14 Linear transformation of DGP ................................................................................... 44
Figure 3.15 Linear transformation of solar radiation .................................................................... 45
Figure 3.16 Linear transformation of UDI and open percent ....................................................... 45
Figure 4.1 The occlusion imaging of base case ............................................................................ 47
Figure 4.2 The occlusion imaging of tilted louvers from 0˚ degrees to 75˚ (from right to left) ... 48
Figure 4.3 Opening percentage of louvers in different tilting degrees ......................................... 48
Figure 4.4 Comparison of changes between opening percentage and tilted angles ...................... 49
Figure 4.5 Base case in Los Angeles in HDR ............................................................................... 49
Figure 4.6 Base case in Seattle in HDR ........................................................................................ 49
Figure 4.7 DGP chart in Los Angeles ........................................................................................... 50
Figure 4.8 DGP chart in Seattle .................................................................................................... 50
Figure 4.9 The illuminance distribution of UDI100-2000 lux in Los Angeles (left) and Seattle
(right) ............................................................................................................................................ 51
Figure 4.10 The solar radiation in Los Angeles (left) and Seattle (right) ..................................... 51
Figure 4.11 False color of HDR of plastic louvers at 12 PM on fall equinox in Los Angeles .... 52
Figure 4.12 False color of HDR of mirror louvers at 12 PM on fall equinox in Los Angeles ..... 52
Figure 4.13 False color of HDR of plastic louvers at 12 PM on winter solstice in Los Angeles 52
Figure 4.14 False color of HDR of mirror louvers at 12 PM on winter solstice in Los Angeles 52
Figure 4.15 DGP of plastic louvers on fall equinox in Los Angeles ............................................ 53
Figure 4.16 DGP of mirror louvers on fall equinox in Los Angeles ............................................ 53
Figure 4.17 DGP of plastic louvers on winter solstice in Los Angeles ........................................ 54
Figure 4.18 DGP of mirror louvers on winter solstice in Los Angeles ........................................ 54
Figure 4.19 False color of HDR of plastic louvers at 12 PM on fall equinox in Seattle .............. 55
Figure 4.20 False color of HDR of mirror louvers at 12 PM on fall equinox in Seattle .............. 55
Figure 4.21 False color of HDR of plastic louvers at 12 PM on winter solstice in Seattle .......... 55
Figure 4.22 False color of HDR of mirror louvers at 12 PM on winter solstice in Seattle .......... 56
Figure 4.23 DGP of plastic louvers on fall equinox in Seattle ..................................................... 56
Figure 4.24 DGP of mirror louvers on fall equinox in Seattle...................................................... 56
Figure 4.25 DGP of plastic louvers on winter solstice in Seattle ................................................. 57
Figure 4.26 DGP of mirror louvers on winter solstice in Seattle.................................................. 57
Figure 4.27 UDI100-2000 lux of plastic louvers in Los Angeles ................................................. 58
pg. 8
Figure 4.28 UDI>2000 lux of plastic louvers in Los Angeles ...................................................... 58
Figure 4.29 UDI100-2000 lux of mirror louvers in Los Angeles ................................................. 59
Figure 4.30 UDI>2000 lux of mirror louvers in Los Angeles ...................................................... 59
Figure 4.31 UDI100-2000 lux of plastic louvers in Seattle .......................................................... 60
Figure 4.32 UDI>2000 lux of plastic louvers in Seattle ............................................................... 60
Figure 4.33 UDI100-2000 lux of mirror louvers in Seattle .......................................................... 60
Figure 4.34 UDI>2000 lux of mirror louvers in Seattle ............................................................... 61
Figure 4.35 Solar radiation in cooling season in Los Angeles ..................................................... 61
Figure 4.36 Solar radiation in cooling season in Seattle ............................................................... 61
Figure 4.37 Solar radiation in cooling season ............................................................................... 62
Figure 4.38 Louvers in Los Angeles and Seattle at 9AM in fall equinox .................................... 63
Figure 4.39 Louvers in Los Angeles and Seattle at 12PM in fall equinox ................................... 63
Figure 4.40 Louvers in Los Angeles and Seattle at 9AM in winter solstice ................................ 65
Figure 4.41 Louvers in Los Angeles and Seattle at 12PM in winter solstice ............................... 65
Figure 4.42 UDI in low and high latitudes ................................................................................... 66
Figure 4.43 Plastic louvers at 12 PM on winter solstice in Los Angeles...................................... 67
Figure 4.44 Mirror louvers at 12 PM on winter solstice in Seattle ............................................... 67
Figure 4.45 UDI of Plastic and mirror louvers in LA ................................................................... 68
Figure 4.46 UDI of Plastic louvers in Los Angeles ...................................................................... 68
Figure 4.47 UDI of Plastic louvers in Seattle ............................................................................... 68
Figure 4.48 UDI100-2000 lux of plastic louvers in Seattle .......................................................... 69
Figure 4.49 UDI100-2000 lux of mirror louvers in Seattle .......................................................... 69
Figure 4.50 UDI of Plastic and mirror louvers in Seattle ............................................................. 70
Figure 5.1 Numerical ranking in Los Angele and Seattle ............................................................. 72
Figure 5.2 Single weighted ranking in Los Angeles and Seattle .................................................. 74
Figure 5.3 Double weighted ranking in Los Angeles and Seattle................................................. 77
Figure 5.4 Ranking of 3D geometry in Los Angeles .................................................................... 79
Figure 5.5 Ranking of 3D geometry in Seattle ............................................................................. 80
Figure 5.6 Qualification of 3D geometry in Los Angeles ............................................................ 81
Figure 5.7 Qualification of 3D geometry in Seattle ...................................................................... 82
Figure 9.1 Plastic base case in Los Angeles ................................................................................. 91
pg. 9
Figure 9.2 Plastic base case in Seattle .......................................................................................... 91
Figure 9.3 Louver in 0 ˚ ................................................................................................................ 92
Figure 9.4 Louver in 15 ˚ .............................................................................................................. 92
Figure 9.5 Louver in 30 ˚ .............................................................................................................. 93
Figure 9.6 Louver in 45 ˚ .............................................................................................................. 93
Figure 9.7 Louver in 60 ˚ .............................................................................................................. 94
Figure 9.8 Louver in 75 ˚ .............................................................................................................. 94
Figure 9.9 Louver in 0 ˚ ................................................................................................................ 95
Figure 9.10 Louver in 15 ˚ ............................................................................................................ 95
Figure 9.11 Louver in 30 ˚ ............................................................................................................ 96
Figure 9.12 Louver in 45 ˚ ............................................................................................................ 96
Figure 9.13 Louver in 60 ˚ ............................................................................................................ 97
Figure 9.14 Louver in 75˚ ............................................................................................................. 97

List of Tables
Table 2.1 Suggestion of the definition of daylight glare comfort classes. Both limits (DGP and
average DGP within 5% band) have to be fulfilled ...................................................................... 25
Table 2.2 The Optimization Parameters ....................................................................................... 28
Table 3.1 Parameters setting in Radiance ..................................................................................... 36
Table 3.2 Parameters in DGP simulation ...................................................................................... 37
Table 3.3 Parameters in UDI simulation....................................................................................... 37
Table 3.3 Parameters in UDI simulation....................................................................................... 37
Table 3.4 Properties of testing materials....................................................................................... 38
Table 3.5 Example of paired comparison ..................................................................................... 46
Table 4.1 Glare comfort class in fall equinox ............................................................................... 63
Table 4.2 Glare comfort class in winter solstice ........................................................................... 66
Table 5.1 Rank of parameters in Los Angeles .............................................................................. 72
Table 5.2 Rank of parameters in Seattle ....................................................................................... 72
Table 5.3 Score of single priority in one parameter in Los Angeles ............................................ 73
Table 5.4 Rank of single priority in one parameter in Los Angeles ............................................. 73
pg. 10
Table 5.5 Score of single priority in one parameter in Seattle...................................................... 73
Table 5.6 Ranking of single priority in one parameter in Seattle ................................................. 74
Table 5.7 Overall ranking of priority in one parameter for architecture ...................................... 74
Table 5.8 Score of priority in two parameters for architecture in Los Angeles ............................ 75
Table 5.9 Ranking of priority in two parameters for architecture in Los Angeles ....................... 75
Table 5.10 Score of priority in two parameters for architecture in Seattle ................................... 75
Table 5.11 Ranking of priority in two parameters for architecture in Seattle .............................. 75
Table 5.12 Score of double priority in one parameter for engineering in Los Angeles ............... 76
Table 5.13 Ranking of single priority in one parameter for engineering in Los Angeles ............ 76
Table 5.14 Score of priority in two parameters for engineering in Seattle ................................... 76
Table 5.15 Ranking of priority in two parameters for engineering in Seattle .............................. 76
Table 5.16 Overall ranking of priority in two parameters ............................................................ 77
Table 5.17 Performance of parameters in terms of architecture viewpoint in Los Angeles ......... 78
Table 5.18 Performance of parameters in terms of architecture viewpoint in Seattle .................. 79
Table 5.19 Performance of parameters in terms of engineering viewpoint in Los Angeles ......... 80
Table 5.20 Performance of parameters in terms of engineering viewpoint in Seattle .................. 81

pg. 11
ABSTRACT
In contemporary architecture, architects are increasingly using exterior solar control screens such
as perforated metal panels and louver systems to reduce solar heat gains for highly glazed
facades. However, the specific geometry of these screens is often based on aesthetics and not on
energy performance or indoor environmental quality considerations. As a consequence, not all of
geometries are capable of providing the level of solar and glare control required and are thus
retrofit with additional façade layers that serve to limit the original daylighting and view
objectives on which the façade was originally based. This is particularly true for 2-dimensional
systems, such as perforated metal screens. Because of this reason, the hypothesis of this research
is that three-dimensional systems can lead to better energy and IEQ performance outcomes
compared with two-dimensional geometry. For early phase design, overall performance of
fenestration not only depends on parameters which include glare discomfort probability, useful
daylight illuminance and solar radiation, but also refers to visual connection to the outdoors
(view). Nevertheless, commonly existing researches did not examine view with other
performance parameters simultaneously, but architects concern the outcome of view more than
the effect of other parameters. Thus, the research examined the geometries in 2D and 3D
dimension with four parameters including view, DGP, UDI and solar radiation. The perforated
screen represented 2D base case and the louvers were 3D tested case. The measurements were
simulated in Los Angeles and Seattle by Ladybug and Honeybee, and the louvers were from 0˚
to 75˚ in matte and specular materials. To develop the concept of design guide, the weighted
score by linear transformation ranked the overall performance of those geometries, and the
paired comparison was used for understanding potential sacrifices. As a result, when the opening
was south-facing, the louvers worked better than the perforated screen. In terms of design
concept, for an architect, the best performance is the horizontal louver in Los Angele and Seattle
except for considering both of view and glare in Seattle, and the exceptional louver is in 45
degrees. For an engineer, the best performance is the louver in 30 degrees in Los Angeles, but
the louver in 45 degrees is better choice in Seattle. The potential sacrifice is glare issue based on
the architecture choice, and part of illuminance maintenance might loss in terms of the
engineering option. Based on the simulated measurement, the overall performance in three
dimensions is better than it in two dimensions, and the results provide concept to develop the
design guide in specific requirements to correspond with environmental conditions.
pg. 12
1 INTRODUCTION
Every façade design is unique interpretation from the architects’ creative thought, and every
idea is based on human’s need and environmental issues. As innovative technologies emerge,
façade design is pushed to a higher level and more complicated construction for great
performance expectations, Complex Fenestration Systems (CFSs) show up as advanced
daylighting systems which refer to window technology, and the main goals of CFSs are to
optimize the luminous environment and improve visual comfort. For instance, shading
devices are combined with the advanced clear window products to make efficient-use of
daylight and to reduce the unwanted solar heat gains and potential glare problem associated
with the clear window product alone. Design of complex geometry is often used to provide a
simultaneous solution for view access, daylight harvesting and solar control, but the relation
between variable geometries in 3 dimensions is still not well examined. The design in this
research is aimed at as a solution of solar control screen in 3 dimensions to develop the
concept guide in the design phase.
In present, even if the design of solar control becomes more complex and aesthetic
performance fulfills with imagination, the core of the intended target is still not fully
achieved for balancing visual solar control and visual view. Sometimes, complex
fenestrations may increase the potential risk of discomfort glare by enhancing the window
luminance, decrease the outdoor/indoor view, and enlarge the illuminance uniformity on
work planes due to forward light scattering. There are two kind of lighting from outdoor: one
is sunlight, one is diffuse skylight and another one is light reflected from neighbor landscape.
For visual comfort, the main goal of fenestration is to block the unnecessary sunlight which
is glare but also transmit the view which is illuminance from skylight and landscape (Yin,
2011a). Also, the geometry of fenestration systems performing this goal well or not is
decided by visual range and environmental conditions, and the visual range is obtained by
eyes viewing up, straight or down in 3 dimensions. Therefore, the percentage of view is
visible for people relying on not only no barrier between their eyes and landscape but also
depending on block between their eyes and sun. Because of this purpose, 3 dimensional solar
control screens have more potential to realize the balancing of solar control and visual view
at the same time, and this research evaluates the geometric parameters in reference to the
measurements of solar control and visual view.
pg. 13
In addition, the materials used in the design are also daylighting performance factors,
because their reflectance and texture affect illuminance level and light distribution (Gregg D.
Ander, 2003). Testing geometry with material features is beneficial to understand and
examine the target models in terms of visual comfort.
1.1 Hypothesis
The louver is the simplified 3D geometry which can be optimized as a solar control
screen and perform better than existing 2D screens in simultaneous terms of visual view,
glare reduction, illuminance level and solar heat radiation.
1.2 Importance
The evaluation of geometric parameters in 3 dimensions will support architects to
develop better solar control based on environmental conditions. To accomplish solar
control by fenestration design, the evaluation of geometry is necessary consideration in
the designing stage.
1.3 Present Problem
Many buildings are famous for their CFSs as a solution to decrease solar heat gain and
glare but also maintain interior illuminance, and architects love to create innovative solar
control screens in the ways of designing fenestration systems in 2 dimensions or 3
dimensions geometry to realize their goals. The great potential of these systems is,
however, not being realized due to barriers to evaluate those complex geometries
associated with solar control and visual comfort capabilities in 3 dimensions, and lack of
appropriate implementations by software simulation in the past. In the geometric design
of solar control screen in 2 dimensions, it can work well as shading for decreasing solar
heat gain and support enough illuminance, such as perforated screen; however, the
limitation of 2D geometry is that it is hard to prevent glare when sunlight is transmitted
through fenestration to human’s eyes at low angles (Figure 1.1).
pg. 14

Figure 1.1 Glare from perforated screen (the De Young museum by Herzog and
Demeuron)
For example, the US Federal Building in San Francisco is famous for the perforated
screen built in the south-east side. The perforated screen provides good shading, but it
failed to solve the glare issue when sun is at low zenith angle.

1.4 External conditions
1.4.1 Orientation/location
In addition to relevant sun-path similarities, different climates cause tglare problems in
the daily or seasonal occurrence of sky patterns, cloudiness, and sunshine duration. A
single set of environmental conditions and solar incident angles enable a simple and
representative calculation. These metrics successfully describe standard fenestration
(NFRC 100, 1997). Experienced architects know that the roughly southeastern opening
has to deploy some protection against excessive glare during sunny conditions. Effective
daylight design depended on correctly understanding the implications of the effects of
solar geometry at specific dates on glazed elements of different orientation at different
latitudes.
pg. 15
1.4.2 Sky Condition
There are three basic types of sky conditions that refer to the percentage of clouds in the
sky, and they are clear, partly cloudy, and overcast. There are many more complicated
sky condition schemes. The Perez sky model defines 240 (Perez, 1987), but the three
prototypes are sufficient for discussion. They define exterior illuminance levels and
distributions. Clear conditions are based on the bright disk of the sun set against a dark
blue sky with high levels of contrast. CIE clear sky is defined by having less than 30% of
clouds covering the sky or no cloud. In either case, the sky is brighter towards the
location of the sun, and the sun is visible. Direct sunlight can be considered and
calculated inside a building. This model is useful when visual glare and thermal
discomfort studies are performed. Overcast sky conditions more evenly distribute light.
Under overcast conditions, this sky with completely cloudy sky (100% covered) model
has been widely used to calculate daylight factor.
1.5 Existing Complex Fenestration Design
1.5.1 Novel louvers
The shape of louvers are also improved for the daylight harvesting function with
responsive control systems, and many types of louvers are on the market. The toothed
blade that improved conventional Venetian blinds by RETRO Lux is one of the new
louvers that provides protection against low-angle winter sunlight and high-angle summer
sunlight, thus avoiding both overheating and glare issues without darkening the interior,
as the blinds are usually kept open (Figure 1.2). The Lightlouver firm takes similar
advantage of geometry for purposeful use of sunlight to meet the illumination
requirements of an architectural space, because their mirror upper side and white lower
side of the louvers reflect the short wave length of the sunlight as well as the sun and the
window glazing (Figure 1.3.).

pg. 16

Figure 1.2 Example and Application of RETRO Lux

Figure 1.3 Example of Lightlouver

1.5.2 Perforated screen
Perforated metal is a common choice for the application of CFSs, as it provides the
longest life of any screening material along with the flexible sizes of hole and spacing to
precisely control the light level, view and privacy level desired. For example, the new
offices for a law firm specializing in the subject in Chicago’s West Loop well presents
perforated screen on the south-facing glass wall, because it is flooded with light from the
split-roof’s south-facing strip skylight and glazed roof section over the courtyard. The
architects also put a circular pattern in the glass to mimic the perforated screen and draw
in a softer light. In addition, there is micro-structural perforated shading screen (MSPSS)
which is made from a stainless steel sheet with elliptical holes smaller than 1 mm and
those holes on the screen are cut in a downward direction to reduce transmission from
sources above the horizon and increase transmission from below the horizon. It is the
advanced technology of perforated design with combination of solar and glare protection
providing direct view out. Figure 1.4 shows the observers’ views of MSPSS: the
pg. 17
observers have clearer view when the distance is greater between viewers and MSPSS. In
fact, it is a step in the direction of a 3 dimensional screen.

Figure 1.4 View through MSPSS (left), unobstructed view (right) (David Appelfeld,
Andrew McNeil and Svend Svendsen 2012)

This research defines the metallic translucent façade as expanded metal mesh, and this
“translucent” façade can be used for controlling solar transmission and preserving
transversal vision due to it three dimensional geometry. The geometry of expanded metal
meshes influences the behavior of radiation, and some angles of incidence will be
reflected and the other will pass through the metal meshes because the incidence of sun
hits curved/sloped metal meshes at different angles (Figure 1.5.). There are several types
of expanded metals which, depending on the form of the blade and the shape of the holes,
such as the rhombus, square and round. The most common type are the rhombus holes.
Besides, the translucence of expanded metal meshes not only acts as a filter but also
provides view for indoor occupants: the curved geometry blocks the upper sun
transmission and people see the city view from the lower side. (Figure 1.6.) For
instance, the New Museum of Contemporary Art of New York’s is famous for the
pg. 18
rhombus-shaped expanded metal mesh. Their metal skin which is partially marked with
windows provide views of New York City and generate various lighting ambiance.

Figure 1.5 Effect of expanded metals on solar transmission and eye’s view

Figure 1.6 New Museum of Contemporary Art of New York’s rhombus-shaped expanded
metal mesh
1.6 Contribution
The result of the study provides the examination and optimization of precisely modeled
form of the solar screen and tunes the screen geometry to specific climate-based
conditions that will be beneficial for architects to design better performing solar controls,
because the results will show appropriate parameters and the limitation of louvers of solar
control screen under specific external conditions. By using Grasshopper’s plug-ins,
Honeybee and Ladybug, to build precise parametric modeling of solar control screens
and performance analysis, the study will show how to develop the simplest geometry of
louvers by parametric process analysis, and what is the key parameter affecting the main
performance in their specific condition to provide the general guideline in early design
stage.
pg. 19
1.7 Study Boundaries
For the selection of CFSs, the solar control screen is the research target of evaluation and
optimization for solar control, and the screen type is the louver systems (Figure 1.7.).
Besides, the only target surface examined in this research is the south orientation in LA
and Seattle.

Figure 1.7 COLT Solarlouver External Solar Shading System at HSE Bootle in UK
The research focuses on adjusting the geometric parameters of solar control screens based
on performance measurements including opening percentage of view, daylight glare
probability, useful daylight illuminance and solar radiation by Grasshopper’s plugins,
Ladybug and Honeybee, which use RADIANCE and DAYSIM as the engines of daylight
simulation. For environmental comfort, it depended on the analysis of daylight glare
probability and view evaluation for visual comfort. Therefore, the parameters of
optimization do not only refer to improved geometric features, such as the size of
apertures, in terms of specific climatic and seasonal factors. The study does not include
analysis of the effect of combination of intelligent control.
For analysis of the effect of daylight harvesting performance, the study will only
concentrate on software simulation by Ladybug and Honeybee to verify the result of
daylight design in existing buildings.
pg. 20
1.7.1 Deliverable
In the early design stage, designers are able to refer to the analysis in this research to
properly improve the geometric parameters based on the louvers and avoid potential
mistakes to grow their ideas in the following process, so the cost of revising the design
will be lower and save a lot of time and effort. On the other hand, this study will
encourage people to apply the similar methods to different geometries of other solar
control screens, so the advanced research of geometry in CFSs will provide more general
guidelines to achieve the goal of high performance façade.
1.7.2 Outlines of the Study
Chapter 1: Introduction of Fenestration Systems in terms of Solar Control
Chapter 2: Background for Evaluation of Solar Control
Chapter 3: Methodology of Simulating Solar Control Screen in the Targeted Types
Chapter 4: Result of Performance Measurements
Chapter 5: Discussion for Ranking
Chapter 6: Conclusion of Optimization from Examination, Performance Analysis and
View Evaluation
Chapter 7: Future Work for simulation of Solar Control Screen in Terms of Geometric
Parameters

pg. 21
2 BACKGROUND RESEARCH
2.1 Case study of an existing building
Design of high performance buildings has become popular, and many buildings
contribute their excellent performance to sustainability. Nevertheless, there is a common
issue neglected when designers conduct their green designs, and the issue is visual
comfort. One of the cases is the U.S. Federal Building in San Francisco (FBSF),
California. This building was designed for sustainability through using natural sources,
such as solar energy. The design team integrated natural ventilation and daylight in the
tower section to minimize the energy consumption. In Kyle Konis’s research (Kyle
Konis, 2013), it mentioned the problem with the south-east façade showing view by
adjusting horizontal tilt of panels for daylighting and solar control strategies. The SE
exterior façade was perforated exterior metal for decreasing direct solar radiation, but this
device failed to effectively prevent glare for occupants because the perforated exterior
metal panels provided deficient glare discomfort control resulting from a direct view of
the solar disc (Figure 2.1). To achieve the daylight level, people often have to sacrifice
their visual comfort because of glare, but occupants in the perimeter zones generally keep
opening a portion of the vision window to maintain outdoor views despite the negative
feedback. This survey recommends that perforated or screen-like exterior shading that
has sufficient depth to effectively block direct radiation and shading control of solar
screens should be -incorporated with consideration of glare discomfort in the early
designing stage to develop a comprehensive evaluation of outcomes of solar control.

pg. 22

Figure 2.1 Example of transmission of direct sunlight and personal modification on SE
façade

2.2 3D Parametric Geometry
In Nancy Cheng’s origami of Shading Light, she was interested in studying the geometry
of sun-shading screens from 2D cutting and folding motifs to 3D forms based on folded
paper. This study created variations of flower petal-like pockets and concentrated on
how tesselating curved folds in 3D patterns through a gridded 6’ x 15’ Shaping Light
Veil installation can show that the relation between digital process and physical test and
exploring parametric surface of structure can adjust lighting level and heat gain in
response to different seasons.
By modulating the curl’s radius of curved folds and lifting a distance between two parts
of the sheet, it is possible to create delicate combinations of cylindrical surfaces and
diffuse incoming sunlight while the sculptural surfaces variably reflect incident light.
This shading project adopted compressed petals and tension petals in seeking the
elegance and efficiency of origami whose sine wave had a corresponding sharp-cornered
zigzag and a linear repeat pattern of alternating concave and convex had a corresponding
zigzag herringbone or chevron patterns, and shadows are cast when diagonal folds
provide canted surfaces which block sunlight. Since the pattern of petals corresponds
sine wave and herringbone Chevron patterns, it is possible to change the proportions of
pg. 23
apertures with Grasshopper software, and daylight transmission may be parametrically
adjusted by spacing and amplitude of chevron patterns. One design direction is to
maximize the opening for light with a constant petal dimension while maintaining enough
of pleated surfaces to create a lattice-like frame.
In their scale model with a south-facing window, the hourly images examined the
compressed and tensioned petals of screen under clear and overcast sky in different
seasons. In summer, the images show that both of the screens block direct sunlight and
reduce solar heat gain and glare; during winter, both screens allow sunlight to deeply
penetrate into the room to take advantage of solar heat gain. In fall or spring, the screens
allow light to enter into the front of the room. Under the overcast sky conditions, the
screen blocks more than half of the incoming light. Moreover, the compressed petals
transmit more light, but the tensioned petal blocks slightly more light and diffuse light
more evenly. The result implies that the tensioned screen more effectively decreases glare
and creates more visually pleasing patterns than the compressed screen.

Figure 2.2 Compressed screen at 3 PM in four seasons (Nancy Cheng, 2012)

Figure 2.3 Tensioned screen at 3 PM in four seasons (Nancy Cheng, 2012)

According to her research, the geometric form can be applied into 3 dimensions with
many possibilities and parametric dimensions can be investigated in terms of sun-shading
pg. 24
screen. The digital modeling in testing can be optimized to better distribute natural light
and protect from glare and heat gain into spaces with specific climatic conditions.
2.3 Daylight Glare Probability
During daytime, the outdoor daylight conditions change due to cloudiness patterns and
the sun position and consequently also glare situations can vary. Daylight glare
probability (DGP), has been developed specifically for the evaluation of daylight glare
issues inside the space. Daylight glare probability (DGP) was developed by Wienold and
Christoffersen (J. Wienold and J. Christoffersen, 2005), and the basic form of equation of
DGP was:

where Eυ is based on the vertical illuminance at the eye [lux], Ls is the luminance of the
source [cd/m2], ωs is the solid angles of the source [sr], and P is the Guth position
index. For a dynamic calculation of glare, there are three methods based on the daylight
glare probability DGP (Wienold and Christoffernsen, 2006):
 Timestep by timestep calculation
 Simplified daylight glare probability DGPs
 Enhanced simplified DGP calculation
Ladybug and Honeybee plugins in Grasshopper use enhanced simplified DGP calculation
where DGPs is based on vertical eye illuminance and simplified images.

The idea is to calculate a simplified image that contains the main glare sources, and the
calculation depends on the vertical illuminance at the eye and the detected glare sources
(size, luminance and position) (Wienold, 2009). The enhanced simplified DGP
calculation is that it is not only time-saving compared to the first method but also more
reliable than second method. At the same time, the façade also shows a direct
pg. 25
transmission component or a peak scattering in sun ray direction as calculating a
simplified image for evaluation with Evalglare provides the correct vertical illuminance,
without spending calculate the exact luminance distribution in the room. It is important to
get the accuracy of rendering the glare sources and to reduce the rendering time, so
selecting proper calculated interreflections is necessary. For non-scattering façade
materials, 60 sun positions are simulated, and the ambient calculation within
RADIANCE (using –ab 0) leads to reliable results. In the investigated methods and
shading devices, the relative root mean squared errors (rRMSE) indicates that increasing
the simulation parameter –ab from 0 to 1 shows no improvement in accuracy (Jan
Wienold, 2009).
Based on user assessments, DGPs shows a very strong correlation with the user’s
response regarding glare perception. For evaluation of the dynamic DGP values, daylight
glare comfort classes are rated by the glare rating classification based on the results from
simulations and frequency distributions. The classes use the upper level of the 95%
confidence intervals of rating scale as DGP limits, and it suggests to restrict the integral
DGP value within 5% interval (Table 2.1) (Jan Wienold, 2009). For example, the Class
B, in 95% of the office time, the DGP must be lower or equal than 95% confidence
interval of “perceptible”.
Table 2.1 Suggestion of the definition of daylight glare comfort classes. Both limits
(DGP and average DGP within 5% band) have to be fulfilled

pg. 26
2.4 Climate-based Daylight Metric, Useful Daylight Illuminance.
Recent surveys found that accurate evaluation of daylight performance can only be
achieved by considering predictions for absolute values of daylight illuminance founded
on realistic meteorological data. Therefore, climate-based daylight modeling is the
prediction of various radiant or luminous quantities, such as irradiance and illuminance,
using sun and sky conditions derived from standard meteorological datasets to represent
the prevailing conditions for occupied interval. Besides, a period of an entire year is
needed to capture the full range in climatic conditions that is based on occupants’
operating period. There are 3 principal modes of climate-based evaluation: cumulative
annual mode, cumulative monthly mode and time-series mode.
Various climate-based daylight metrics have been formulated in the late 1990s, and
useful daylight illuminance (UDI) which is one of daylight metrics, combines time-step
and location factors. UDI is defined as the annual occurrence of illuminances across the
work plane that are within a range considered “useful” by occupants. The “useful” range
is based on occupant preferences and behavior in daylit offices with shading devices
operated by users. For non-residential buildings, the range of daylight illuminance
between 500 lux to 2000 lux or 2500 lux is desirable. It is believed that the glare issue is
the common problem in non-residential buildings, such as office buildings, so the upper
limit might be 2000 lux instead of 2500 lux. Generally, UDI-autonomous (UDI-a) in the
range 500 to 2500 lux will need less artificial lighting for occupants. The UDI scheme is
applied by the following calculation at the point of the occurrence of daylight levels for
non-residential buildings (LBNL-4585E, 2009):
• The illuminance is less than 100 lux, i.e. UDI `fell-short' (or UDI-f).
• The illuminance is greater than 100 lux and less than 500 lux, i.e. UDI supplementary
(or UDI-s).
• The illuminance is greater than 500 lux and less than 2,500 lux, i.e. UDI autonomous
(or UDI-a).
• The illuminance is greater than 2,500 lux, i.e. UDI exceeded (or UDI-e)
pg. 27
2.5 Climate-based Daylight Metric, Spatial Daylight Autonomy
Spatial Daylight Autonomy “is a metric describing annual sufficiency of ambient
daylight level in interior environments. It is defined as the percent of an analysis area that
meets a minimum daylight illuminance level for a specified fraction of the operating
hours per year.” (LEED v4) sDA is the parameter to measure illuminance –sufficiency
which calculates the percentage of analysis points that exceed a specified illuminance
level (300 lux) for at least 50% of the total occupied hours between 8 a.m. and 6 p.m.
over the year (IES, 2012). According to the new LEED v4 rating system, two points can
be gained if the sDA value is achievable in 55% of regularly occupied space, and three
points can be gained if the sDA value is achievable in 75% of occupied space. The first
difference between sDA and UDI is that sDA is only based on the minimum illuminance
level except for the maximum limit, and the reported hours above 300 lux at least for
50% of that time. Moreover, it is expressed as the percentage of the area in a room
meeting the specific requirement. The numbers are more directly related to the built
geometry. The sDA calculation grids should be no more than 2 feet (600 millimeters)
square and laid out across the regularly occupied area at a work plane height of 30 inches
(76 millimeters) above the finished floor (unless otherwise defined). Use an hourly time-
step analysis based on typical meteorological year data or an equivalent, for the nearest
available weather station.
In the research for optimizing façade driven by daylighting (Y. Elghazi et al., 2014), it
presented that the hybrid double façade in the shape of the Kaleidocycle rings (Figure
2.4) showed good potential to improve daylighting performance through incorporating
simulation tools and genetic optimization with a parametric design.
pg. 28

Figure 2.4 The rotation of unit configuration and different rotation angles form 0 to 90 degrees
(Y. Elghazi et al., 2014)

It used DIVA which contains RADIANCE and DAYSIM as the daylight engine, and the
simulation logic follows the Approved Method for IES Spatial Daylight Autonomy
(sDA) and Annual Sunlight Exposure (ASE). Those metrics connect building occupants
with the outdoors and reduces the use of electrical lighting by introducing daylight into the
space. The measurements of daylight availability are rated to 3 levels including day lit,
partially lit and over lit, and geometric parameters are rotation angles and opening sizes
(Table 2.2). The simulation process is done in two stages: the first stage showed
potentials in increasing daylight performance by changing façade typology, and the
objective in the second stage was set to maximize the daylighting area and to minimize
the over lit area and ASE areas.

Table 2.2 The Optimization Parameters
Optimization Parameters
Opening Size 20 cm to 65 cm (step 1 cm)
Rotating Angle 0 ˚ to 90 ˚ (step 1 ˚)

The model was a south facing façade in Cairo, Egypt, and the result indicated that
Kaleidocycle rings of 30 cm size and 64˚ rotating angle reached daylighting distribution
that is better than LEED v4 requirements and passes Daylight Availability standards.
pg. 29

Figure 2.5 The optimization of Kaleirdocycle configuration in daylight performance (Y.
Elghazi et al., 2014)

2.6 Solar Radiation
Façade transparency of buildings directly not only affected the occupant’s visual comfort
but also impacted their thermal comfort. High fenestration ratio usually leads to high
cooling demand and overheating in addition to glare problem. For example, the studies
showed that 70% the peak cooling load of a building in Frankfurt, Germany, comes from
excessive solar heat transmitting through the glazing façade (Kuhn, T., 2006). In Sweden,
the climate was colder and less solar radiation, but a study concluded that energy
performance strongly related to the size of glazing of the building envelopes (Poirazis, H.
and Blometerberg, A., 2008). Therefore, examination of solar transmission was an
important factor to evaluate the solar control. The survey (Alan Pino et al., 2012)
researched 228 cases of office building in Santiago of Chile (33˚S) where the weather
was warm and dry with high solar radiation in the entire year, and the daily temperature
difference was large in January. This study used four variables were modified: Window-
to-wall ratio (WWR), lack of or type of external solar protection devices, glazing type
and orientation. The Figure 2.6 showed the relation between four orientations and solar
radiation.
pg. 30

Figure 2.6 Temperature evolution inside offices during a winter day with clear sky
conditions and external dry bulb temperature from 0 ◦C to 20 ◦C, for a building with 50%
WWR (Alan Pino et al., 2012)

The most influential parameter on energy demand was the WWR. Independent of glazing
and the solar protection type used, a building with low WWR (20%), would obtain a
demand lower than 40 kWh/m
2
year. With 50% WWR, the energy total energy would
fluctuate between 40 and 70 kWh/m
2
year. Finally, for a 100% WWR would fluctuate
between 50 and 155 kWh/m
2
year. Higher WWR increased energy demand variability
because it was less insulated and allows higher solar gains than concrete with external
insulation. If sample means, grouped by solar protection type, were compared with cases
without solar protection, it was observed that this strategy can reduce energy demands in
around 30% and could reduce variability of the samples values. However, the WWR was
the dominant variable. A significant advantage could not be noticed between the two
types of solar protection studied.
pg. 31

Among the studied cases the most efficient corresponds to the 20% WWR building, with
selective double glazing, North oriented overhang (OH), solar protection, and horizontal
blinds (HB) on East and West oriented windows. It generates a thermal status which
registers no significant differences between different oriented offices. In cooling periods
the most demanding office is the North- West-oriented with 19.6 kWh/m
2
year, and the
less demanding is South-oriented with 13.9 kWh/m
2
year. It concludes that that there
exists summer overheating in buildings having 100% WWR and important variability on
energy demand is produced if using large glazed areas.
2.7 Material Properties
Material properties from reflecting surfaces will significantly impact daylight
performance and the reflectance of materials that affect direction of the incident light.
The reflectance is the ratio of reflected flux to incident flux, and it is referred as the
direct/indirect reflectance (Nick Backer and Koen Steemers, 2002). Specular and
diffusing materials are delouvered by the processes when light falls onto a surface. For an
opaque surface, it means that the reflected light beams remain concentrated in a bundle
upon leaving the surface when light incident falls on smooth surface, such as mirrors. On
the other hand, if the surface is rough, such as the asphalt road, the light rays will reflect
and diffuse in many different directions (Figure 2.8).
Figure 2.7 Box plot of total annual energy demand with respect to use of
solar protection devices (Alan Pino et al., 2012)
pg. 32

Figure 2.8 Reflection process of specular and diffuse materials

In simulation tools, accurate information on the materials used for internal surfaces is also
fundamental. For example, RADIANCE-simulation show different distributions of
internal luminance views and visual impression RADIANCE-simulation luminance views
with a minor change in specularity (Figure 2.9).

Figure 2.9 Simulated view of a room with walls made of diffuse oak (specularity 0.0%),
varnished checkmate oak (specularity 0.2%) and varnished shining oak (specularity
2.3%) (M. Bodart et al, 2007)

pg. 33
3 METHODOLOGY
To examine the performance of different geometries, perforated screen is used as a base case
to represent 2D geometry where the U.S. Federal Building in San Francisco has the problem
with glare. Louvers with different geometries are the test cases which are adopted to
examine the hypothesis: the fenestration performance of 3D geometry is better than the base
case in two dimensions. With the results from a computer-aided simulation tool, the overall
performance is examined, using two proposed evaluation methods, the weighted scores and
pair comparison, which are based on the selected key performance parameters, including
opening percentage, daylight glare probability, useful daylight illuminance and solar
radiation (referring to Chapter 4). Grasshopper and Ladybug were used to examine and
evaluate geometric effects. Grasshopper is used for creating 3D parametric design in this
research, and the simulation was processed by the simulation engines, Radiance and
EnergyPlus, with assistance of Ladybug and Honeybee. . Integration of the parametric tools
in Grasshopper provides feedback on design modifications in almost real time, as it runs
within interaction between the design condition and the environmental information. Besides
performance parameters, the open percentage is adapted to qualify the view from the room
towards the outside. The test condition is the 3 dimensional louver and the reference
condition is a perforated 2 dimensional screen. The opening is south-facing, and Los
Angeles and Seattle are selected as test locations. Metric units were used in the tests due to
the requirements of Ladybug and Honeybee. The model is a 3m x 10m x 3m (W x L x H)
shoebox. (Figure 3.1). The sky condition is sunny with the sun (Clear sky). The studies are
conducted with one geometric parameter at a time in order to keep the inputs separate.
pg. 34

Figure 3.1 Size of shoebox in modeling

3.1 Modeling Tools
3.1.1 Ladybug and Honeybee
Ladybug and Honeybee are used to import climate-based conditions to Grasshopper and
thus provide a variety of 2D and 3D interactive graphics to support the decision-making
process during the initial stages of design. They simplify the process of daylighting and
energy analysis and provide easily understood visualizations in the 3D modeling interface
of Rhino/Grasshopper. In addition, users are able to work with validated energy and
daylighting engines such as EnergyPlus, Radiance and Daysim. Moreover, these tools are
open source, which makes it possible for users to customize them based on specific
needs. In addition, Radiance is a suite of programs for the analysis and visualization of
lighting in design.
Ladybug is a tool for environmental analysis in a single parametric platform. It supports
interactive 2D and 3D visualization for weather data and those results provide useful
information in the initial designing stage.
pg. 35

Figure 3.2 Components in Ladybug
Similar to Ladybug, Honeybee is designed to run analyses on building masses but for
more advanced studies. The workflow is designed for designers, and many of the
parameters can be assigned with customized values, instead of merely default values.

Figure 3.3 Component in Honeybee
There are four main steps for Ladybug and Honeybee
1) Preparing simulation geometry
Users can set the percentage of openings for each orientation, which is used to
calculate and add the openings to the geometry for energy and daylighting simulation.
2) Check the input file:
Ladybug provides a way (the import/export connection) that users can import back
the simulation file and visualize it in the Rhino/Grasshopper environment before the
simulation process.
3) Run the simulation(s):
For daylighting simulation, the user needs to provide test surfaces or test points and
overwrite radiance in details, such as number of bounces, sampling, and etc. A path
connected to the weather file is also needed as well as a working directory and project
name. By default, Honeybee uses identical geometries for both energy and
daylighting studies, and extracts material properties for daylighting simulation from
the EnergyPlus construction (Figure 3.3).
4) Visualize the results:
pg. 36
Honeybee can re-import and visualize the energy and daylighting simulation results,
which maps the results with the geometries and analyzes them. For example,
components in DAYSIM calculate an annual climate-based daylighting analysis, such
as Daylight Autonomy. Furthermore, users can customize parameters, such as the
period of study or range of thresholds. Other components import and visualize the
hourly results for users to access real time examination (Figure 3.4.).

Figure 3.4 Components for real time examination
3.1.2 Rendering in Radiance
Radiance adopts distributed ray tracing for rendering, which is powerful to simulate the
diffuse light distribution and reflection in a three dimensional environment. Distribution
ray tracing adopts Monte Carlo integration to solve the rendering equation. This
technique was introduced by Cook et. Al. (Cook, 2010) . It is notable because of its
simplicity and ability to simulate area luminaires, camera lens effects and imperfect
specular reflection. There are many parameters in Radiance to calculate accurate
simulation of lighting action, and this research uses ambient bounces (ab), ambient
accuracy (aa), ambient resolution (ar), ambient divisions (ad) and ambient super-samples
(as) in glare and illuminance simulation. Table 3.2 and table 3.3 show the value used in
simulation of DGP and UDI in this research. For “min” value, it shows the fastest and
roughest rendering, however it is not necessarily the smallest value numerically. The
"fast" value gives a reasonably fast rendering. The "accur" value gives a relatively
accurate rendering. The "max" value gives the ultimate in accuracy.
Table 3.1 Parameters setting in Radiance
Parameter Description Min Fast Accur Max
ab ambient bounces 0 0 2 8
pg. 37
aa ambient accuracy 0.5 0.2 0.15 0
ar ambient resolution 8 32 128 0
ad ambient divisions 0 32 512 0
as ambient super-samples 0 32 256 1024

Table 3.2 Parameters in DGP simulation
Parameter Description Value
ab ambient bounces 2
aa ambient accuracy 0.15
ar ambient resolution 128
ad ambient divisions 32
as ambient super-samples 256

Table 3.3 Parameters in UDI simulation
Parameter Description Value
ab ambient bounces 2
aa ambient accuracy 0.5
ar ambient resolution 8
ad ambient divisions 0
as ambient super-samples 0

Table 3.4 Parameters in UDI simulation
Parameter Description Value
dc direct certainty 0.25
(Fast)
dp direct pretest density 64
(Fast)

False color images are an alternate representation of pixel data for HDR images and they
are utilized to reveal HDR lighting data, which cannot be displayed in absolute values
and full range through simulations. In those images, a range of colors is assigned to a
range of luminance values, and it visualizes the spatial distributions within a scene.
pg. 38

Figure 3.5 False color HDR image of plastic louver in 15˚ at 12 PM on winter
solstice in Los Angeles
3.2 Variables
There are three variables in the test. The first variable is location, which represents the
low and high latitudes. Los Angeles and Seattle are selected. The second variable is
time: the time includes 9AM/12 PM and the date contains fall equinox and winter
solstice. The third variable is material, and mirror and plastic material are adopted. For
materials, the measurement is simulated by the optical properties. The mirror metal and
plastic are selected to be as specular and matte materials (Table 3.4).
Table 3.5 Properties of testing materials
Red
Reflectance
Green
Reflectance
Blue
Reflectance
Specularity

Roughness
Mirror metal 1 1 1 1 0
Plastic 0.35 0.35 0.35 0 0.05

3.3 Geometry
3.3.1 Base Case
The base case is a perforated screen, and the diameter of each circle is 20 cm. It is used to
compare with parametric louvers by the following parameters (Figure 3.6). The grid size
of measurement is 5cm.
pg. 39

Figure 3.6 Geometry of base case
3.3.2 Louvers
The louver is the test model to examine geometric effects on daylight performance and
view quality. Tilted angle is examined under constant spacing between louvers for this
research. It evaluates each parameter separately to identify which parameter is more
influential than the others or which one is more important in specific ambiance
conditions. For the variable angles, the tilting degrees are from 0 ˚, 15˚, 25˚, 35˚, 45˚, 55˚,
65˚ and 75˚ while the depth of louvers and the spacing are still 20 cm each (Figure 3.7
and Figure 3.6).
pg. 40

Figure 3.7 Geometry of louvers

Figure 3.8 Titling angles downward from 0 ˚ to 75 ˚

3.4 Key parameters
3.4.1 View
Opening percentage is calculated from occlusion ratio, which qualifies the view from
indoors. The opening percentage is the transmission ratio of the outdoor view, and the
sum of transmission ratio and occlusion ratio is one, which means that the opening
percentage is that one minuses the occlusion ratio.
Opening Percentage = 1- Occlusion Ratio
The occlusion ratio of a view is defined to be the fraction of pixels of the view that are
hidden by other objects, such as louvers, when projected on the camera sensor. The
camera will be set in the center of the shoebox to capture those views of testing surfaces
from indoors and create images of those surfaces. The image is transformed into black
pg. 41
and white by Photoshop. Black means the testing surface is blocking the view (Figure
3.9). Then black ratio is calculated from ratio of black pixels which present blocked area.

Figure 3.9 Way of calculating occlusion ratio
3.4.2 Daylight glare probability
Daylight glare probability (DGP) is used for the qualification of geometric parameters of
the fenestration system. HDR imaging for DGP is provided by Ladybug and Honeybee.
The distance between shading surface and observing position is 2 meters, with a fisheye
lens lined up with the center on the vertical surface, and the simulation type is luminance
in medium quality (Figure 3.10).

Figure 3.10 Observing point of DGP in shoebox
The simulation of DGP on winter solstice (12/21) and fall equinox (9/23) are processed at
9AM and12 PM in 2014. The observing point is at 2 meters deep and 1.5 meters in height
at the center of shoebox. For DGP evaluation, this research compares tilted angles and
depth of louvers.
pg. 42
3.4.3 Useful daylight illuminance
In this research, the UDI level is applied by calculation of Honeybee and the occupied
period is from 9 a.m. to 5 p.m. with an hour of lunch break:
• The illuminance is less than 100 lux.
• The illuminance is greater than 100 lux and less than 200 lux.
• The illuminance is greater than 2,000 lux, i.e. UDI exceeded (or UDI-e).(Futrell,
Ozelkan, & Ph, 2013)
The evaluated surface is in 1 meter height for UDI (Figure 3.11).

Figure 3.11 Measured surface for UDI in shoebox
After simulating UDI, the testing surface shows illuminance levels where UDI is above
2000 lux and UDI between 100 lux to 2000 lux.
3.4.4 Solar Radiation
To achieve good thermal performance, solar radiation plays an important role in interior
conditioned zones. As the fenestration area becomes larger, daylighting performance and
visual view could be possibly enhanced; however, larger opening brings higher solar
radiation into the indoor space and increases the cooling load at the same time. Therefore,
to conclude the balanced performance from geometries, it is essential to consider solar
loads while developing features of different geometries. The measurement of solar
radiation is to use the Ladybug_RadianceAnalysis component (Figure 3.12) in Ladybug,
and the analysis period is in the cooling season from July to September. It is used for
pg. 43
examining the thermal performance in addition to illuminance level and glare issue. The
solar radiation is measured on vertical surface where the distance between testing
geometry and measured surface is 10 cm, and grid size on the vertical surface is 5 cm
(Figure 3.13).

Figure 3.12 Component of Radiation Analysis

Figure 3.13 Vertical surface measuring solar radiation

3.5 Comparison of Measurements
The geometries are examined for four parameters, occlusion ratio, DGP, UDI and solar
radiation to distinguish the difference between matte and specular materials at two
locations, Los Angeles and Seattle matte and specular materials are examined for glare
issues at different time. For DGP, the DGP of plastic and mirror louvers are simulated on
fall equinox and winter solstice, the results show DGP from one material of louvers at 9
pg. 44
AM and 12 PM one time. For example, the result from the measurements of the plastic
louvers in Los Angeles could reveal DGP at 9AM and 12 AM in fall equinox in a chart.
For UDI, the images show illuminance distribution in the desired illuminating over
illuminated area, and in terms of the material features, UDI would be compared in the
same chart at one location. For solar radiation, the simulation results at two locations in
cooling season, July 1
st
to September 30
st
shows at once. In addition, the evaluation of
latitude effect would be based on the plastic louvers.
3.6 Performance Evaluation
To verify the hypothesis, this research normalizes the values of four parameters by linear
transformation. In linear transformation, the ranges of values of those parameters are
transferred into the same range, and the total points are one hundred which is equally
divided into each parameter. For example, the maximum point of DGP is 25, which
presents the best performance. Because of recommended classes of DGP, the geometry
can get 25 points when DGP is not above 0.3. In other words, that geometry gets zero
point if DGP reaches 1. Therefore, a linear transformation for DGP is the function
(Figure 3.14). This research visualizes this function by its graph, which is a line through
𝑇 (𝑥 ) = −
250
7
𝑥 +
250
7
the origin with slope −
250
7
and the value of slope is negative
because the high DGP means high discomfort from glare. Similarly, the solar radiation
has a negative correlation with performance in terms of cooling season, so its equation is
calculated in the same way. In addition, since the solar radiation is 201 kWh/m
2
when
the open percent is 100%, that value would be considered as the worst value for the
lowest point of original range (Figure 3.15). In other words, 201 kWh/m
2
get zero in the
adjusted scale.

Figure 3.14 Linear transformation of DGP

pg. 45

Figure 3.15 Linear transformation of solar radiation
However, higher UDI and opening percentage present better performance, and both of
their original ranges are from 0% to 100%. Thus, the linear transformation are the same
as with each other (Figure 3.16).

Figure 3.16 Linear transformation of UDI and open percent
In addition, to effectively examine the advantages and disadvantages of geometry with
specific requirements, the paired comparison would be adopted, because this approach
help people understand not only benefits but also sacrifices between different choices. In
chapter 5, it would adopt the following steps of the paired comparison:
1. Make a list of four options for parameters. Assign each parameters with different
points from 0 to 25 based on adjusted scores from linear transformation in the
previous step.
2. Mark those parameters as both the row and column headings on the worksheet to
compare options with one-another (Table 3.5).
3. Within each of the blank cells, compare the parameter in the row with the option
in the column to decide which of the two options is most important.
4. Write down the points of the most important option in the cell. Then, score the
difference in importance between the options, running from zero (no difference)
to three (one much more important than the other.) (Table 3.5).
pg. 46
Table 3.6 Example of paired comparison
View (V) DGP (G) UDI (U)
Solar
Radiation
(S)
Score
Percentage
View (V) V3 I2 V1 4
36%
DGP (G) G2 R1 2
18%
UDI (U) R2 2
18%
Solar Radiation (S) 3
28%
Total 11
100%

5. Finally, consolidate the results by adding up the values for each of the options and
convert these value into percentage of total score.
6. Weighted the scores by specific preference. For an architects’ preference, view is
3, glare is 2, UDI is 1, and solar radiation is 0.5. From an engineer’s view, solar
radiation is 3, UDI is 2, and the others are 1.

pg. 47
4 RESULTS OF MESUREMENTS
4.1 Opening percentage
Because of the nature of human vision, the shape of a perspective projection is a frustum
which is the part of a conical solid left after cutting off the top portion with a plane
parallel to the base, and the position of viewpoint affects occluded objects in visual
views. Compared with parallel projection, perspective projection is able to simulate the
3D view in terms of human vision. The result of opening percentage for perforated base
case was 58% open (Figure 4.1).

Figure 4.1 The occlusion imaging of base case
Interestingly, the upper and middle views were gradually occluded by louvers when tiled
angles rose; however, the lower view was less occluded when louvers tilted from 0
degree to 30 degrees. Also, the view was increasingly occluded after tilted angles were
over 45 degrees (Figure 4.2), and the opening percentage decreased when tilting angles
increased. Although the tilting constantly increased 15 degrees each time, the changes of
opening percentage for visual view increased until 45 degrees and decreased after that
angle (Figure 4.4). Compared with the ratio of perforated base case, the ratio of the
louver in 30 degrees was the closest one. The higher occlusion ratio also represented a
larger visible view for people.
pg. 48

Figure 4.2 The occlusion imaging of tilted louvers from 0˚ degrees to 75˚ (from right to
left)

Figure 4.3 Opening percentage of louvers in different tilting degrees

65%
59%
48%
32%
20%
10%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
Open percecnt
Tilting Dgrees
0˚ to
15˚
15˚
to
30˚
30˚
to
45˚
45˚
to
60˚
60˚
to
75˚
Change of percent open 7% 11% 15% 12% 11%
Change of degrees 15 15 15 15 15
0
5
10
15
20
25
30
35
40
45
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
Change of open percent
Change of tilted angles ˚
pg. 49
Figure 4.4 Comparison of changes between opening percentage and tilted angles

4.2 Base case
For DGP, the base case of the perforated screen in Los Angeles, at 9 AM, DGP was
around 0.5 in both the fall equinox and the winter solstice for plastic and mirror louvers.
Compared with louvers examined at 9 AM, DGP, at 12 PM, increased 7% in fall for all of
the louvers, and it increased largely to 100% in winter. The DGP of the base case in
Seattle was similar with Los Angeles except for DGP at 9AM in winter, because DGP
decreased to 35% for plastic louvers and 36% for mirror louvers. At 12 PM in winter
solstice, the result clearly showed that the perforated screen needs to block the glare in
both of Los Angeles and Seattle; similarly, San Francisco Federal Building whose façade
adopts perforated screen faces the same issue. According to Figure 4.5 and Figure 4.6, the
perforated screen lead to glare problem at 12PM on winter solstice, and that is the issue
this research would like to improve by 3D geometry.

Figure 4.5 Base case in Los Angeles in HDR

Figure 4.6 Base case in Seattle in HDR
pg. 50

Figure 4.7 DGP chart in Los Angeles

Figure 4.8 DGP chart in Seattle

For UDI, the UDI was 62% in Los Angeles, and it was 53% in Seattle. Although the
pattern of illuminance distribution looked alike, the intensities in Los Angeles are higher
than it in Seattle (Figure. 4.6). For solar radiation, it was 116 kWh/m
2
in the cooling
0.49
0.56
0.52
0.59
0.5
1
0.52
1
9 12
DGP
TIME
Base Case in Los Angeles
Fall, plastic Fall, mirror Winter, plastic Winter, mirror
0.49
0.56
0.52
0.59
0.35
1
0.36
1
9 12
DGP
TIME
Base Case in Seattle
Fall, plastic Fall, mirror Winter, plastic Winter, mirror
pg. 51
season in Los Angeles, and it was 142 kWh/m
2
in Seattle. The area in high radiation was
larger when the perforated screen was in Seattle (Figure 4.7).

Figure 4.9 The illuminance distribution of UDI100-2000 lux in Los Angeles (left)
and Seattle (right)

Figure 4.10 The solar radiation in Los Angeles (left) and Seattle (right)

4.3 Louvers
4.3.1 Daylight Glare Probability
DGP of louvers in different tilted angles was examined by false color HDR images. The
measurements in Los Angeles and in Seattle showed firstly, and the results revealed the
data from 9AM and 12PM on fall equinox on winter solstice separately. Besides, the
louvers tested in the simulation used plastic and mirror for material comparison.
In Los Angeles, DGP at 12PM was higher than DGP at 9AM in all conditions, and
specular material was a higher DGP than matte material. On the fall equinox, DGP
steadily decreased from the lowest to the highest tilting degrees. On winter solstice, DGP
gradually decreased with increasing tilting degrees except for tilting degrees in 15
pg. 52
degrees, and DGP reached to 100% when plastic and mirror louvers were tilted
downward to 15 degrees. The result showed that glare discomfort reaches to the worst
conditions when the tilted angle was 15 degrees at 12 PM on winter solstice, and that was
not only because of lower solar altitude but also due to the position of the view point that
led to some solar beams transmitting to the indoors. For a good class of DGP, DGP was
below 0.4. For plastic louvers, the tilted angles should be over 30 degrees to achieve good
class of DGP; on the other hand, for mirror louvers, the tilted angles should be over 45
degrees to fit in the same class.

Figure 4.11 False color of HDR of plastic louvers at 12 PM on fall equinox in
Los Angeles

Figure 4.12 False color of HDR of mirror louvers at 12 PM on fall equinox in
Los Angeles

Figure 4.13 False color of HDR of plastic louvers at 12 PM on winter solstice
in Los Angeles

Figure 4.14 False color of HDR of mirror louvers at 12 PM on winter solstice
in Los Angeles
pg. 53

Figure 4.15 DGP of plastic louvers on fall equinox in Los Angeles

Figure 4.16 DGP of mirror louvers on fall equinox in Los Angeles
0.51
0.44
0.4
0.35
0.27
0.28
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Fall equinox: : plastic louver in LA
9AM
12PM
0.6
0.52
0.54
0.49
0.34
0.26
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Fall equinox: : mirror louvers in LA
9AM
12PM
pg. 54

Figure 4.17 DGP of plastic louvers on winter solstice in Los Angeles

Figure 4.18 DGP of mirror louvers on winter solstice in Los Angeles

In Seattle, the measurements were at 9AM/12PM on fall equinox and winter solstice, and
plastic and mirror louvers were tested. Compared with DGP in LA, although DGP, on fall
equinox, decreased when the tilted angles from the lowest to the highest degrees, DGP in
0.6
1
0.44
0.39
0.32
0.26
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Winter solstice: plastic louvers in LA
9AM
12PM
0.66
1
0.57
0.52
0.37
0.27
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Winter solstice: mirror louver in LA
9AM
12PM
pg. 55
Seattle were generally higher them in LA. Besides, DGP slightly increased when mirror
louvers were tilting downward 30 degrees. On winter solstice, DGP decreased from 0
degree to 75 degrees, and DGP were higher than them on fall equinox. At 12PM, DGP of
plastic and mirror louvers reached to 100% when tilted angles between 0 to 30 degrees,
and DGP deeply decreased when the louvers were tilted downward over 45 degrees. For
plastic louvers, the tilted angles should be over 30 degrees to achieve good class of DGP;
on the other hand, for mirror louvers, the tilted angles should be over 45 degrees to fit in
the same class.

Figure 4.19 False color of HDR of plastic louvers at 12 PM on fall equinox in
Seattle

Figure 4.20 False color of HDR of mirror louvers at 12 PM on fall equinox in
Seattle

Figure 4.21 False color of HDR of plastic louvers at 12 PM on winter solstice
in Seattle
pg. 56

Figure 4.22 False color of HDR of mirror louvers at 12 PM on winter solstice in Seattle

Figure 4.23 DGP of plastic louvers on fall equinox in Seattle

Figure 4.24 DGP of mirror louvers on fall equinox in Seattle

0.56
0.47
0.41
0.37
0.3
0.27
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Fall equinox: : plastic louvers in Seattle
9AM
12PM
0.6
0.52
0.54
0.49
0.34
0.26
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Fall equinox: : mirror louvers in LA
9AM
12PM
pg. 57

Figure 4.25 DGP of plastic louvers on winter solstice in Seattle

Figure 4.26 DGP of mirror louvers on winter solstice in Seattle

4.3.2 Useful Daylight Illuminance
The useful daylight illuminance is in the range from 100 lux to 2000 lux in a year. The
illuminance distributions of UDI showed in Los Angeles, and it compared the plastic and
mirror louvers firstly. Then, the measurement of plastic and mirror louvers showed in
Seattle later.
1 1 1
0.37
0.3
0.265
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Winter solstice: plastic louvers in Seattle
9AM
12PM
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degree
Winter solstice: mirror louvers in Seattle
9AM
12PM
pg. 58
In Los Angeles, images of illuminance distribution compare UDI100-2000 lux and UDI>2000 lux
between different materials. For plastic louvers, the bottom of this image was the opening
side, and the depth of UDI100-2000 lux area decreased by the increasing tilted angles from the
left image to the right image. The large the angle, the deeper the space for UDI100-2000 lux,
and there was higher percentage of UDI above 2000 lux when the area was close to the
window.
Figure 4.27 UDI100-2000 lux of plastic louvers in Los Angeles
Figure 4.28 UDI>2000 lux of plastic louvers in Los Angeles
pg. 59
Figure 4.29 UDI100-2000 lux of mirror louvers in Los Angeles

Figure 4.30 UDI>2000 lux of mirror louvers in Los Angeles

In Seattle, images of illuminance distribution compare UDI100-2000 lux and UDI>2000 lux for
different materials. For plastic and mirror louvers in Seattle, the pattern of useful
illuminance distribution is comparable with the illuminance distribution in Los Angeles.
In Seattle, the difference of UDI between the two locations was the intensity of UDI100-2000
lux in the room was lower. Briefly, the value of UDI100-2000 lux and the UDI>2000 lux were lower
for Seattle than for Los Angeles.
pg. 60

Figure 4.31 UDI100-2000 lux of plastic louvers in Seattle

Figure 4.32 UDI>2000 lux of plastic louvers in Seattle

Figure 4.33 UDI100-2000 lux of mirror louvers in Seattle

pg. 61

Figure 4.34 UDI>2000 lux of mirror louvers in Seattle

4.3.3 Solar radiation
The only variable in solar radiation is locations. The distributions of solar radiation are
similar to each other in Los Angeles and Seattle. In Los Angeles and Seattle, the solar
radiation in the cooling season decreases when tilted angles were higher. The solar
radiation in Los Angeles was lower than it was in Seattle and the difference between the
two locations were beginning to converge within increasing angles of louvers.

Figure 4.35 Solar radiation in cooling season in Los Angeles

Figure 4.36 Solar radiation in cooling season in Seattle

0
10
20
30
40
50
60
0 15 30 45 60 75
Solar Radiation (kWh/m
2
)
Tilting Degree
Solar radiation in cooling season
LA
Seatle
pg. 62

Figure 4.37 Solar radiation in cooling season

4.4 Effect of latitude on DGP and UDI
4.4.1 Evaluation of latitude effect on DGP
Different values of DGP can be referred to suggestion of the daylight glare comfort
classes, the DGP of matte louvers were classified into 3 classes including best class (A),
good class (B) and reasonable class (C). In addition, the result was evaluated by the
factor of latitudes at 9 AM and 12 PM in fall equinox and winter solstice. Los Angeles
represented a lower latitude compared with Seattle in high latitude. In fall equinox, the
DGP were similar at 9 AM, and that means tilted angles had less effect on latitude
difference because people would feel the same level of glare discomfort at this time. At
12 PM, the difference between two locations rose, and it led to the maximum difference
of DGP increasing to 5% when the louvers were horizontal. In addition, the DGP in
Seattle were higher than in Los Angeles in all tilted angles except for the smallest tilting
degrees.
In Los Angeles, when the tilting angles above 45 degrees, the glare comfort achieved
class A at 9 AM and 12 PM. The glare discomfort was intolerable when the louvers were
horizontal at 12 PM. In Seattle, the glare comfort achieved class A at 9 AM when the
louvers were above 30 degrees, but, at 12 PM, the angles of louvers should tilt downward
more than 60 degrees to reach class A. Glare discomfort was intolerable when tilting
degrees were 0 or 15 at 12 PM (Table 4.1).
pg. 63

Figure 4.38 Louvers in Los Angeles and Seattle at 9AM in fall equinox

Figure 4.39 Louvers in Los Angeles and Seattle at 12PM in fall equinox

Table 4.1 Glare comfort class in fall equinox
Location Time Tilting Degrees
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Louvers in low and high latitude at 9 AM
LA
Seatle
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Louvers in low and high latitudes at 12 PM
LA
Seatle
pg. 64
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
Los Angeles
9 AM C B B A A A
12 PM intolerable C B A A A
Seattle
9 AM C B A A A A
12 PM intolerable intolerable C B A A

In winter solstice, the DGP in Seattle achieve Class A at 9 AM when tilted angles were
above 0 degrees, and at 12 PM, DGP rapidly increased and became intolerable. The
results implied that DGP in low latitude are more sensitive to change of time. The largest
difference of DGP was 56% when the tilting degrees was 30 and the second maximum
was 40% when the louvers were horizontal. In addition, the DGP values in Seattle were
higher than those in Los Angeles, and the increase of DGP on winter solstice was higher
than those on fall equinox. In Los Angeles, the worst glare discomfort happened when the
louvers were in 15 degrees; in Seattle, the worst case showed when louvers were from 0
degree to 30 degrees.
In Los Angeles, when the tilting degrees were above 45 degrees, the glare comfort could
achieve class A at 9 AM, and louvers need to tilt to 60 degrees to maintain the A class at
12 PM. For the horizontal louver, the glare discomfort was intolerable both in the
morning and at noon. In Seattle, the glare comfort could achieve class A at 9 AM except
for horizontal louvers, but, at 12 PM, the louvers in the same angle only fit in Class A
when louvers kept the minimum opening. Glare discomfort was intolerable when tilting
degrees were from 0 degree to 45 degrees at 12 PM (Table 4.2).
pg. 65

Figure 4.40 Louvers in Los Angeles and Seattle at 9AM in winter solstice

Figure 4.41 Louvers in Los Angeles and Seattle at 12PM in winter solstice

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Louvers in low and high latitude at 9 AM in winter
solstice
LA
Seatle
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP
Tilting Degrees
Louvers in low and high latitudes at 12 PM
LA
Seatle
pg. 66
Table 4.2 Glare comfort class in winter solstice
Location Time
Tilting Degrees
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
Los Angeles
9 AM
intolerable C B A A A
12 PM
intolerable intolerable C B A A
Seattle
9 AM
C A A A A A
12 PM
intolerable intolerable intolerable intolerable B A

4.4.2 Evaluation of latitude effect on UDI
In Los Angeles and Seattle, both UDI values gradually decreased when the tilted angles
gradually increased. When angles of louvers were between 0 degrees and 45 degrees, the
UDI decreased within the increasing angles in Los Angeles and Seattle. However, UDI
rapidly increased by 35% when 60 degrees of tilted angles and sharply reduced to 2%
when louvers tilted downward an additional 15 degrees in Seattle; on the other hand,
UDI, in Los Angeles, kept decreasing till 75 degrees of tilted angles.

Figure 4.42 UDI in low and high latitudes
4.5 Effect of material effect on DGP and UDI
4.5.1 Evaluation of material effect on DGP
There was more glare discomfort in the room at 12PM on winter solstice, and this
research compared material effects on DGP at this time. In Los Angeles and Seattle,
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
UDI
Tilting Degrees
UDI in Los Angeles and Seattle
LA
Seatle
pg. 67
mirror louvers brought more glare than the plastic louvers. In false color of HDR images,
the shape of sunlight were clearer and the area in red and yellow were larger in mirror
cases.

Figure 4.43 Plastic louvers at 12 PM on winter solstice in Los Angeles

Figure 4.44 Mirror louvers at 12 PM on winter solstice in Seattle

4.5.2 Evaluation of material effect on UDI
In Los Angeles, the depth of UDI100-2000 lux in the room decreased when downward tilting
angles increased. The depth of UDI of mirror louvers slightly reduced, and the UDI of the
louvers in 45 degrees largely rose. The UDI100-2000 lux distribution for mirror louvers was
less predictable, as the angles of louvers were not directly related to the depth of UDI100-
2000 lux. For example, in the room with horizontal louvers, UDI in the 80% of whole room
could reach to the range between 100 lux and 2000 lux. For the louvers at 15 and 30
degrees, illuminance in the rear of the room decreased and UDI100-2000 lux was lower than
40%. For the mirror louvers in 15 degrees, the illuminance distribution are similar, but
UDI100-2000 lux at the rear wall was a little bit higher than it in 30 degrees. When the louver
was in 45 degrees, there was larger area in UDI100-2000 lux even including the area next to
the opening, and there was more light reflected to the rear wall due to high reflectance of
the specular materials. Even if the mirror louvers tilted downwards to 60 degrees, one-
third of the room maintained UDI100-2000 lux. Finally, UDI100-2000 lux deeply decreased when
the louvers were at 75 degrees.

pg. 68

Figure 4.45 UDI of Plastic and mirror louvers in LA

Figure 4.46 UDI of Plastic louvers in Los Angeles

Figure 4.47 UDI of Plastic louvers in Seattle
In Seattle, the change of UDI were similar with the way showing in Los Angeles, but the
intensity of UDI100-2000 lux was lower than it in Los Angeles. UDI100-2000 lux of plastic louvers
0 15 30 45 60 75
Plastic 71.55 63.68 50.39 30.29 14.35 3.97
Mirror 72.56 56.32 52.69 73.88 48.27 3.98
0
10
20
30
40
50
60
70
80
100 lux < UDI< 2000 lux
UDI: Plastic and mirror louvers in LA
pg. 69
kept decreasing to 21% with the decreasing angles. When the angle of plastic louver was
75 degrees, UDI100-2000 lux reached 2%. For mirror louvers, UDI100-2000 lux gently decreased
when tilted angles were between 0 to 30 degrees. When tilted angle was 45 degrees,
UDI100-2000 lux increased to 21%. When angles tilting was downward 60 degrees, UDI100-2000
lux decreased 22%, and it largely decreased to 3% when tilted angle was in 75 degrees.
For material properties, plastic louvers performed better than mirror louvers did when the
tilted angles were between 0 degree and 15 degrees. Mirror had higher UDI100-2000 lux
when the tilting degrees were over 30 degrees, and the difference of UDI100-2000 lux reached
the highest difference when the tilted angle was in 45 degrees.

Figure 4.48 UDI100-2000 lux of plastic louvers in Seattle

Figure 4.49 UDI100-2000 lux of mirror louvers in Seattle

pg. 70

Figure 4.50 UDI of Plastic and mirror louvers in Seattle

0˚ 15˚ 30˚ 45˚ 60˚ 75˚
Plastic 61% 53% 40% 21% 9% 2%
Mirror 60% 49% 47% 61% 39% 3%
0%
10%
20%
30%
40%
50%
60%
70%
UDI
UDI: Plastic and mirror louverd in Seattle
pg. 71
5 EVALUATION OF OVERALL PERFORMANCE
The hypothesis of this research is that the overall performance in 3D geometry is better
than it in 2D geometry. To examine the overall performance of geometry, the four
parameters were given a specific score (referring to Chapter 3.6). The sum of scores for
each geometry represents the overall performance, and the ranking of performance is
based on their scores. For specific preference, the type of rank was based on numerical
score, single weighted score and double weighted scores. In the end, the results revealed
that perforated screen performed worse than the louvers either in Los Angeles or in
Seattle. Compared with the louvers, the base case did poor in prevention from glare
issues and solar radiation, although the UDI and opening percentage reached to the
average level.
5.1 Ranking of Geometries by linear transformation
For the single numerical score, the linear transformation adjusted each parameter to the
range of 0 to 25 points, and higher points mean higher performance. Since DGP and solar
radiation were undesirable, the slopes of transformation were negative. Conversely, the
view and UDI are desirable, so the slopes are positive (referring to chapter 3.6).
5.1.1 Ranking of single numerical score
The numerical score was the sum of scores from four parameters without any weighting.
Because glare discomfort happened most frequently at 12 PM on winter solstice, they are
used for evaluating the performance as the worst cases. In the first step of linear
transformation, the influence of each parameter was divided equally, and it showed
overall performance for the perforated screen and the tilted louvers in Los Angeles and
Seattle (Table 5.1 and Table 5.2). In Los Angeles, the horizontal louver performed better
than the others, since its opening percentage and UDI were well ahead of the others even
if its performance of DGP and solar radiation did not work that well. The second best
choice was the louvers at 30 degrees, which was just one point lower than the horizontal
louvers. The worst case was the base case, because it not only failed to block glare but
also could not prevent solar radiation. In Seattle, the louver tilting downward 45 degrees
was the best performance, as it successfully block glare and solar radiation. The second
choice is the louver at 60 degrees, although it also worked well in prevention of glare and
pg. 72
solar radiation. The case within louvers in 45 degrees also lost the points in the UDI and
open percent. The worst choice was still perforated screen, because it still lost many
points on prevention from glare and solar radiation.
Table 5.1 Rank of parameters in Los Angeles

Base
Case
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP 0 14 0 20 22 24 25
UDI 16 18 16 13 8 4 1
Open Percent 11 16 15 12 8 5 2
Solar Radiation 11 19 21 23 24 24 24
Total Points 37 68 52 67 61 57 53
Rank 7 1 6 2 3 4 5

Table 5.2 Rank of parameters in Seattle

Base
case
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
DGP 0.0 0.0 0.0 0.0 22.5 25.0 25.0
UDI
13 15 13 10 5 2 1
Open Percent
11 16 15 12 8 5 2
Solar Radiation
7 18 21 23 24 24 24
Total Points 31 50 49 44 59 56 53
Rank 7 4 5 6 1 2 3

Figure 5.1 Numerical ranking in Los Angele and Seattle

pg. 73
5.1.2 Ranking of single parameter by weighted score
The single weighted score was for a strongly specific preference, so only the score of the
selected parameter was double weighted and the other parameters kept their original
score, and ranking of geometries referred to their total scores (Table 5.3 to Table 5.6) in
each corresponding parameter. For the base case, its performance was the poorest in Los
Angeles and Seattle. For the louvers, in Los Angeles, the best choice is the horizontal
louver for view or illuminance, and the best choice for glare or solar radiation is the
louver at 30 degrees (Table 5.7). In Seattle, the best choice is the louver at 45 degrees for
any single parameter.

Table 5.3 Score of single priority in one parameter in Los Angeles
PRIORITY BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW 49 83 67 80 70 62 54
GLARE 38 81 52 88 84 81 77
ILLUMINANCE 54 85 68 81 70 61 53
SOLAR RADIATION 49 86 73 91 86 81 76

Table 5.4 Rank of single priority in one parameter in Los Angeles
BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW PRIORITY 7 1 4 2 3 5 6
GLARE 6 3 5 1 2 3 4
ILLUMINANCE 6 1 4 2 3 5 7
SOLAR RADIATION 6 2 5 1 2 3 4

Table 5.5 Score of single priority in one parameter in Seattle
BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW PRIORITY 42 65 64 57 68 61 54
GLARE 31 49 49 45 82 81 77
ILLUMINANCE 44 64 62 55 65 58 53
SOLAR RADIATION 38 67 70 68 84 80 76

pg. 74
Table 5.6 Ranking of single priority in one parameter in Seattle

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW PRIORITY 7 2 3 5 1 4 6
GLARE 6 4 4 5 1 2 3
ILLUMINANCE 7 2 3 5 1 4 6
SOLAR RADIATION 7 6 4 5 1 2 3

Table 5.7 Overall ranking of priority in one parameter for architecture
SINGLE
PARAMETER
VIEW GLARE ILLUMINANCE SOLAR RADIATION
LOS ANGELES 0˚ 30˚ 0˚ 30˚
SEATTLE 45˚ 45˚ 45˚ 45˚

Figure 5.2 Single weighted ranking in Los Angeles and Seattle

5.1.3 Ranking of double parameters by weighted score
In reality, the performance of geometry often depends on more than one parameter.
Thus, this variation picked two parameters as weighted performance that were weighted
differently based on the requirements. From the viewpoint of architecture, architects are
more concerned with view than the other parameters, so the view was triple weighted. It
could be combined with one of the other parameters that was double weighted, and the
weighted scores were added into the original score of the other parameters to examine the
overall performance (Table 5.8 and Table 5.10). For the viewpoint of energy, solar
radiation in the cooling season is directly related to energy consumption, so the solar
pg. 75
radiation was triple weighted. Similarly, the weighted scores were combined with other
parameters again (Table 5.12 and Table 5.14). For the architecture requirement, the
horizontal louvers won the best performance in Los Angeles and Seattle except for one
situation, and the exception is the louver in 45 degrees for view and glare parameters in
Seattle. For energy concerns, the louver in 30 degrees is the best choice for Los Angeles,
and the louvers at 45 degrees work well in Seattle.
Table 5.8 Score of priority in two parameters for architecture in Los Angeles

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW & GLARE 60 113 82 112 100 91 81
VIEW &
ILLUMINANCE
76 117 98 105 86 71 57
VIEW & SOLAR
RADIATION
49 80 61 69 54 43 32

Table 5.9 Ranking of priority in two parameters for architecture in Los Angeles

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW & GLARE 7 1 5 2 3 4 6
VIEW &
ILLUMINANCE
5 1 3 2 4 6 7
VIEW & SOLAR
RADIATION
5 1 3 2 4 6 7

Table 5.10 Score of priority in two parameters for architecture in Seattle

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
VIEW & GLARE 53 81 79 69 98 91 81
VIEW &
ILLUMINANCE
66 96 92 79 81 68 57
VIEW & SOLAR
RADIATION
46 63 58 46 52 42 32

Table 5.11 Ranking of priority in two parameters for architecture in Seattle

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
pg. 76
VIEW & GLARE 6 3 4 5 1 2 3
VIEW &
ILLUMINANCE
6 1 2 4 3 5 7
VIEW & SOLAR
RADIATION
4 1 2 4 3 5 6

Table 5.12 Score of double priority in one parameter for engineering in Los Angeles

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
SOLAR & VIEW 71 121 109 126 118 110 102
SOLAR & GLARE 60 119 94 134 132 129 125
SOLAR &
ILLUMINANCE
60 119 94 134 132 129 125

Table 5.13 Ranking of single priority in one parameter for engineering in Los Angeles

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
SOLAR & VIEW 7 2 5 1 3 4 6
SOLAR & GLARE 7 5 6 1 2 3 4
SOLAR &
ILLUMINANCE
7 5 6 1 2 3 4

Table 5.14 Score of priority in two parameters for engineering in Seattle

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
SOLAR & VIEW 56 101 106 103 115.5 109 102
SOLAR & GLARE 45 85 91 91 130 129 125
SOLAR &
ILLUMINANCE
45 85 91 91 130 129 125

Table 5.15 Ranking of priority in two parameters for engineering in Seattle

BASE
CASE
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
SOLAR & VIEW 7 6 3 4 1 2 5
SOLAR & GLARE 6 5 4 4 1 2 3
SOLAR &
ILLUMINANCE
6 5 4 4 1 2 3

pg. 77
Table 5.16 Overall ranking of priority in two parameters
APPLICATION ARCHITECTURE ENGINEERING
DOUBLE
PARAMETERS
VG VI VS SV SG SI
LOS ANGELES 0˚ 0˚ 0˚ 30˚ 30˚ 30˚
SEATTLE 45˚ 0˚ 0˚ 45˚ 45˚ 45˚
Note: V is View, G is glare, I is illuminance and S is solar radiation. If V is the former letter, it means the letter has highest
weight than the second one.

Figure 5.3 Double weighted ranking in Los Angeles and Seattle

5.2 Performance examination by paired comparison
It is not only important to find the best performance but is also essential to understand the
potential sacrifice that results. Therefore, the paired comparison was adopted to examine
the louvers that were 3D geometries to understand the benefits and sacrifices after overall
performance was evaluated by double weighted score. To distinguish the potential
sacrifice, the paired comparison also weighted parameters depending on the application,
and compared two selected parameters simultaneously (referring to Chapter 3.6). The
opening percentage which presented the visual view was weighted three times, and DGP
pg. 78
was weighted 2 times while the other parameters kept were kept at one. On the other
hand, in terms of an engineer’s viewpoint, solar radiation and UDI would be related to
potentiality of energy-saving. Thus, the solar radiation was weighted 3 times and UDI
was weighted 2 times while the other parameters kept in one.
5.2.1 Performance examination based on viewpoint of architecture
In Los Angeles, the best tilted angle was 0 degrees, because it achieved the largest view.
Even if the horizontal louvers can work better than the others, it also has a potential glare
problem because of low performance on glare. Therefore, the second choice should be in
30 degrees, because it balanced visual view and glare protection. Besides, the thermal
performance at 30 degrees was good enough compared with the other louvers. Even if the
tilted angles were above 45 degrees and they performed well in glare and heat protection,
the visual view values were too low to choose them.
In Seattle, the best tilted angle is at 0 degree for most of time, but the louver at 45 degrees
worked better when considering view and glare. Indeed, the horizontal louver could not
effectively prevent from glare discomfort especially with sun at low solar altitudes. On
the other hand, although the louver at 45 degrees provided great glare protection, the
performance of view also largely decreased. In fact, it seemed harder to balance visual
view and glare protection at the same time for those louvers in the location in higher
latitude, since great view and glare protection reacted in opposite ways.
Table 5.17 Performance of parameters in terms of architecture viewpoint in Los Angeles
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
OP 80% 69% 38% 16% 8% 4%
DGP 20% 0% 51% 69% 71% 80%
UDI 0% 12% 0% 0% 0% 0%
SR 1% 19% 11% 15% 20% 16%
TOTAL 100% 100% 100% 100% 100% 100%
Note: OP is open percent, and SR is solar radiation.
pg. 79

Figure 5.4 Ranking of 3D geometry in Los Angeles

Table 5.18 Performance of parameters in terms of architecture viewpoint in Seattle
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
OP 76% 70% 62% 16% 9% 4%
DGP 0% 0% 0% 68% 71% 71%
UDI 10% 9% 8% 0% 0% 0%
SR 14% 21% 29% 15% 20% 25%
OVERALL 100% 100% 100% 100% 100% 100%

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0˚ 15˚ 30˚ 45˚ 60˚ 70˚
Performance of each paramters
Titling Degrees
Visual View
Glare Protection
Illuminance Performance
Thermal Performance
pg. 80

Figure 5.5 Ranking of 3D geometry in Seattle

5.3 Performance examination based on viewpoint of energy
In Los Angeles, the best tilted angle was 30 degrees, because it achieved the highest
thermal protection and maintained part of UDI to save potential energy in lighting.
Besides, when the tilted angles were over 30 degrees, they might be an option for an
engineer, as they provided enough thermal protection and saved some electricity from
lighting simultaneously. However, the glare protection was relatively poorer than the
others.
In Seattle, the best tilted angle was at 45 degrees, because it had the best solar protection
and blocked glare at the same time. Unlike the louvers in Los Angeles, those selections
might sacrifice useful daylight illuminance.
Table 5.19 Performance of parameters in terms of engineering viewpoint in Los Angeles
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
OP 1% 7% 0% 0% 0% 0%
DGP 0% 0% 2% 10% 17% 19%
UDI 27% 24% 11% 4% 1% 0%
SR 71% 69% 87% 86% 82% 81%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0˚ 15˚ 30˚ 45˚ 60˚ 70˚
Perfromance of each parameters
Titling Degrees
Visual View
Glare Protection
Illuminance Performance
Thermal Performance
pg. 81
OVERALL 100% 100% 100% 100% 100% 100%

Figure 5.6 Qualification of 3D geometry in Los Angeles
Table 5.20 Performance of parameters in terms of engineering viewpoint in Seattle
0˚ 15˚ 30˚ 45˚ 60˚ 75˚
OP 9% 7% 6% 0% 0% 1%
DGP 0% 0% 0% 13% 18% 19%
UDI 25% 19% 13% 1% 0% 0%
SR 66% 74% 81% 86% 81% 80%
OVERALL 100% 100% 100% 100% 100% 100%

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0˚ 15˚ 30˚ 45˚ 60˚ 70˚
Pefrormance of each parameters
Titling Degrees
Visual View
Glare Protection
Illuminance Performance
Thermal Performance
pg. 82

Figure 5.7 Qualification of 3D geometry in Seattle
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
0˚ 15˚ 30˚ 45˚ 60˚ 70˚
Performance of each parameters
Titling Degrees
Visual View
Glare Protection
Illuminance Performance
Thermal Performance
pg. 83
6 CONCLUSION
For developing the concept of a design guide, this research not only concluded the
traditional parameters including DGP, UDI and solar radiation, but also added visual
view as a parameter to rank the 2D and 3D geometry for the assessment of overall
performance. Based on overall performance, the 2D geometry, perforated screen, was the
base case which was compared with the 3D geometry, tilted louvers in the south-facing
facade. As the result, the overall performance of the base case was worse than the
performance of tilted louvers either in Los Angeles or in Seattle, which was in accord
with the hypothesis in this research. For evaluation of DGP, the time which was at 12 PM
on winter solstice needed to give high priority due to high frequency of glare discomfort.
Besides, although specular material brought more useful illuminance, it also led to higher
glare discomfort. For evaluation of opening percentage, it decreased within the
increasing tilted angles, but the change of opening percentages were different even if
increment of angles were the same in each case. For solar radiation, all of them became
lower when tilted angles rose. Besides, concerning the glare issue, the research suggests
plastic material is better than the specular one, because glare discomfort happened more
frequently when the louvers were specular material.
In terms of specific preference, such as view priority, the tilting degree at 0 degree might
be a better choice in most environments including high latitudes, such as Seattle.
However, if the designer considers view and glare issues at the same time, the louver in
45 degrees is better to balance overall performance (referring Table 5.16). Moreover,
those louvers also might sacrifice glare protection to some extent. For engineers, it is
good for energy performance when the tilting degree is 30 degree when it is located in
low latitudes, and the louver in 45 degrees works better in high latitude. The potential
sacrifice might be the useful illuminance for saving electricity from lighting.
The biggest challenge was to distinguish the proper choice based on certain conditions.
Those four parameters were from different metrics, so the first thing is to normalize them
in the same standard. In addition, the four parameters were related to specific preference;
for example, architects might consider view and glare more than solar radiation. Based on
certain preferences, this research adopted linear transformation to rank the overall
pg. 84
performance and used the paired comparison to understand potential sacrifices. Those
approaches were useful for weighing up the relative importance of different options for
performance evaluation. When the scores of are metrics weighted differently, the results
clearly showed the advantages and disadvantages for certain requirements, and this way
also allowed designers to examine trade-offs easily. Based on the workflow in this
research, the process is adaptable to different solar orientations and climates to develop
the completed concept of geometry design.
pg. 85
7 FUTURE WORK
In this research, the orientation was to the south, so testing the rest of the orientations
would be the first step of future work. Because the locations only contained Los Angeles
and Seattle, other locations, such as Miami, can be included in the future to make
completed evaluations of latitude effects. In addition to the examination of tilted angles,
the aspect ratios are also helpful to assess 3D geometry in details. Louver is the simple
geometry in three dimensions, and it might be good to examine the complex geometry,
such as expanded metal, to accomplish the concept of guide in early design phase. For the
material issue, the louver can use mirror and plastic materials on each side. For
optimization, it may combine with computer programming to get the optimized
geometry.

pg. 86
8 BIBLIOGRAPHY
Angeles, Los. 2013. “Investigation of Evalglare Software , Daylight Glare Probability and High Dynamic
Range Imaging for Daylight Glare Analysis,” 450–63.
Appelfeld, David, Andrew McNeil, and Svend Svendsen. 2012. “An Hourly Based Performance
Comparison of an Integrated Micro-Structural Perforated Shading Screen with Standard
Shading Systems.” Energy and Buildings 50 (July). Elsevier B.V.: 166–76.
doi:10.1016/j.enbuild.2012.03.038.
Bustamante, Waldo, Sergio Vera, Alejandro Prieto, and Claudio Vásquez. 2014. “Solar and
Lighting Transmission through Complex Fenestration Systems of Office Buildings in a
Warm and Dry Climate of Chile.” Sustainability 6 (5): 2786–2801. doi:10.3390/su6052786.
Cheng, Nancy Yen-wen. n.d. “Designing with Material , Visual and Digital Considerations” 2:
603–10.
Cooling, Solar. 2011. “Quality Assurance and Support Measures for Solar Cooling.”
Drgas, Szymon, and Adam Dabrowski. 2012. “Scoring Methods for Normalized Kernels for
Multi-Level Speaker Verification.” 2012 International Conference on Signals and
Electronic Systems, ICSES 2012 - The Conference Proceedings.
doi:10.1109/ICSES.2012.6382251.
Environment, Indoor, Council Canada, Sustainable Buildings, and Natural Resources. 2007.
“Complex Fenestration Systems : Towards Product Ratings for Indoor Environment
Quality” 2: 109–22.
Feb, Jan, Mar Apr, May Jun, Jul Aug, Sep Oct, and Nov Dec. n.d. “Electric Consumption
( kWh ) Gas Consumption ( Btu ).” Design, 5200–5200.
Futrell, Benjamin, Leed Ap, and Dale Brentrup. n.d. “Evaluating Annual Daylighting
Performance Through Statistical Analysis And Graphs : The Daylighting Scorecard.”
Futrell, Benjamin J, Ertunga C Ozelkan, and D Ph. 2013. “Optimizing Building Design for
Daylighting,” 3739–48.
Inanici, Mehlika, and D Ph. 2005. “E RNEST O RLANDO L AWRENCE Per-Pixel Lighting
Data Analysis Per-Pixel Lighting Data Analysis,” no. August.
Inanici, Mn; Kumaragurubaran V. 2012. “High Dynamic Range Image Processing Toolkit for
Lighting Simulations and Analysis Viswanathan Kumaragurubaran A Thesis Submitted in
Partial Fulfillment of the Requirements for the Degree of Master of Science in
Architecture , Design Computing University O.”
pg. 87
Ito, K., and S. Sakane. 2001. “Robust View-Based Visual Tracking with Detection of
Occlusions.” Proceedings 2001 ICRA. IEEE International Conference on Robotics and
Automation (Cat. No.01CH37164) 2. doi:10.1109/ROBOT.2001.932775.
Jakubiec, J Alstan, and Christoph F Reinhart. 2013. “Predicting Visual Comfort Conditions in a
Large Daylit Space Based on Long-Term Occupant Evaluations : A Field Study.” 13th
Conference of International Building Performance Simulation Association, 3408–15.
Jakubiec, J., and C. Reinhart. 2012. “The ‘Adaptive Zone’ - A Concept for Assessing Discomfort
Glare throughout Daylit Spaces.” Lighting Research and Technology 44: 149–70.
doi:10.1177/1477153511420097.
Johansson, Thomas. 2014. “Gendered Spaces : The Gym Culture and the Construction of
Gender,” 32–47.
Killingsworth, By Brian, and Pavement Structures. 2011. “Concrete Sustainability Report,” no.
July: 11–12.
Kittler, Richard, Miroslav Kocifaj, and Stanislav Darula. 2012a. “Daylight Science and
Daylighting Technology” 0. New York, NY: Springer New York: 127–54. doi:10.1007/978-
1-4419-8816-4.
Kittler, Richard., Stanislav Darula, Miroslav Kocifaj. 2012b. “Daylight Science and Daylighting
Technology.” New York, NY: Springer New York. doi:10.1007/978-1-4419-8816-4.
Konis, Kyle Stas. 2011. Effective Daylighting: Evaluating Daylighting Performance in the San
Francisco Federal Building from the Perspective of Building Occupants. UC Berkeley:
Architecture. https://escholarship.org/uc/item/0vd3q8s2
Konis, Kyle. 2013. Evaluating daylighting effectiveness and occupant visual comfort in a side-lit
open-plan office building in San Francisco, California.
Building and Environment 59 (2013) 662e677
Kuhn, Tilmann E. 2006. “Solar Control: Comparsion of Two New Systems with the State of the
Art on the Basis of a New General Evaluation Method for Facades with Venetian Blinds or
Other Solar Control Systems.” Energy and Buildings 38 (6): 661–72.
doi:10.1016/j.enbuild.2005.10.001.
Kwak, Suha, Woonhyun Nam, Bohyung Han, and Joon Hee Han. 2011. “Learning Occlusion
with Likelihoods for Visual Tracking.” Proceedings of the IEEE International Conference
on Computer Vision, 1551–58. doi:10.1109/ICCV.2011.6126414.
March, R P Leslie, L C Radetsky Bs, and A M Smith Ms. 2012. “Conceptual Design Metrics for
Daylighting,” no. May 2011: 277–90.
pg. 88
Mardaljevic, J, M Andersen, N Roy, and J Christoffersen. 2012. “Daylighting Metrics: Is There a
Relation Between Useful Daylight Illuminance and Daylight Glare Probability?” Ibpsa-
England Bso12, 189–96.
Mardaljevic, J, L Heschong, and E Lee. 2009. “Daylight Metrics and Energy Savings.” Lighting
Research and Technology 41 (3): 261–83. doi:10.1177/1477153509339703.
Mardaljevic, John. n.d. “Climate-Based Daylight Analysis for Residential Buildings Impact of
Various Window Configurations , External Obstructions , Orientations and Location on
Useful Daylight Illuminance.”
Mardaljevic, John, Queens Building, and The Gateway. 2006. “Examples of Climate-Based
Daylight Modelling Solar Access Study : The Arts Students League , New York , USA,” no.
67.
Marshall, Jennifer L, Patrick Williams, Jean-philippe Rheault, Travis Prochaska, Richard D
Allen, and D L Depoy. n.d. “Characterization of the Reflectivity of Various Black
Materials,” 1–8.
Multivariate, Modelling, and Extreme Value. 2014. “Biometrika Trust” 77 (3): 245–53.
Nayar, Shree, Ravi Ramamoorthi, and Pat Hanrahan. n.d. “Basic Principles of Surface
Reflectance.”
Ok, Vildan. 1992. “A Procedure for Calculating Cooling Load due to Solar Radiation: The
Shading Effects from Adjacent or Nearby Buildings.” Energy and Buildings 19: 11–20.
doi:10.1016/0378-7788(92)90032-C.
Pino, Alan, Waldo Bustamante, Rodrigo Escobar, and Felipe Encinas Pino. 2012. “Thermal and
Lighting Behavior of Office Buildings in Santiago of Chile.” Energy and Buildings 47
(April). Elsevier B.V.: 441–49. doi:10.1016/j.enbuild.2011.12.016.
Reflectance, Specular, and The Spectralight. n.d. “Daylighting Fact Sheet Daylight That Goes
the Distance.”
Reinhart, Christoph F., and Jan Wienold. 2011. “The Daylighting Dashboard - A Simulation-
Based Design Analysis for Daylit Spaces.” Building and Environment 46 (August): 386–96.
doi:10.1016/j.buildenv.2010.08.001.
Reinhart, Christoph, Tarek Rakha, and Dan Weissman. 2014. “Predicting the Daylit Area - A
Comparison of Students Assess- Ments and Simulations at Eleven Schools of Architecture.”
Reinhart, F, Christoph., John. Mardaljevic, and Zack. Rogers. 2006. “Dynamic Daylight
Performance Metrics for Sustainable Building Design.” L E U K O S V O L 3 N O 1, 7–31.
doi:10.1582/LEUKOS.2006.03.01.001.
pg. 89
Rockcastle, Siobhan, and Marilynne Andersen. 2013. “Annual Dynamics of Daylight Variability
and Contrast: A Simulation-Based Approach to Quantifying Visual Effects in Architecture,”
9–23. doi:10.1007/978-1-4471-5233-0.
Schulze, Stephanie, Jillian M Hughes, and M a Candidate. 2003. “Perceptual Mapping by
Multidimensional By,” no. September.
Shan, Rudai. 2014. “Optimization for Whole Building Energy Simulation Method in Façade
Design,” 1–9.
Smith, Adrian, and Gordon Gill. n.d. “LADYBUG : A PARAMETRIC ENVIRONMENTAL
PLUGIN FOR GRASSHOPPER TO HELP DESIGNERS CREATE AN
ENVIRONMENTALLY-CONSCIOUS DESIGN Mostapha Sadeghipour Roudsari ,
Michelle Pak.”
Suk, Jae Yong., “Post-Treatment Analysis of The Glare Remediation of The Walt Disney
Concert Hall” (PhD diss., University of Southern California, 2007).
Sunguard, Guardian. n.d. “Technical Guidelines.”
Thanachareonkit, Anothai, Eleanor S. Lee, and Andrew McNeil. 2014. “Empirical Assessment
of a Prismatic Daylight-Redirecting Window Film in a Full-Scale Office Testbed.” Leukos
10: 19–45. doi:10.1080/15502724.2014.837345.
Traskos, Steffani. 2013. “‘ Gendered Spaces : The Gym and the Restaurant .’”
Tucker, Tyler. n.d. “PERFORMATIVE SHADING DESIGN,” 1–4.
Tzempelikos, Athanassios, and Andreas K. Athienitis. 2007. “The Impact of Shading Design and
Control on Building Cooling and Lighting Demand.” Solar Energy 81 (3): 369–82.
doi:10.1016/j.solener.2006.06.015.
Wienold, J., Christoffersen, J. (2005a). "Towards a new daylight glare rating," Proceedings of
Lux Europa 2005 - Lumière pour l'homme, Berlin, 19-21 september 2005.
Wienold, J., Christoffersen, J. (2006). "Evaluation methods and development of a new glare
prediction model for daylight environments with the use of CCD cameras," Energy and
Buildings, vol. 38, no. 7, pp. 743-757.
Wienold, Jan. 2009. “Fraunhofer Institute for Solar Energy Systems , Heidenhofstr . 2 , D-79110
Freiburg , Germany Email : Jan.wienold@ise.fraunhofer.de TIMESTEP BY TIMESTEP
CALCULATION – RADIANCE,” 944–51.
Wu, Geman. 2012. “Studies in Preliminary Design of Fenestration : Balancing Daylight
Harvesting and Energy Consumption,” no. November: 3–8.
pg. 90
Zeigler, B. P. 1982. “Simulation Methodology.” Simulation 38 (5): 181–82.
doi:10.1177/003754978203800506.

pg. 91
9 APPENDIX
9.1 False color of HDR imaging in base case

Figure 9.1 Plastic base case in Los Angeles

Figure 9.2 Plastic base case in Seattle

pg. 92
9.2 False color of HDR imaging in louvers within variable angles
Los Angeles

Figure 9.3 Louver in 0 ˚

Figure 9.4 Louver in 15 ˚
pg. 93

Figure 9.5 Louver in 30 ˚

Figure 9.6 Louver in 45 ˚
pg. 94

Figure 9.7 Louver in 60 ˚

Figure 9.8 Louver in 75 ˚
Louvers in Seattle
pg. 95

Figure 9.9 Louver in 0 ˚

Figure 9.10 Louver in 15 ˚
pg. 96

Figure 9.11 Louver in 30 ˚

Figure 9.12 Louver in 45 ˚
pg. 97

Figure 9.13 Louver in 60 ˚

Figure 9.14 Louver in 75˚

Abstract (if available)

Abstract In contemporary architecture, architects are increasingly using exterior solar control screens such as perforated metal panels and louver systems to reduce solar heat gains for highly glazed facades. However, the specific geometry of these screens is often based on aesthetics and not on energy performance or indoor environmental quality considerations. As a consequence, not all of geometries are capable of providing the level of solar and glare control required and are thus retrofit with additional façade layers that serve to limit the original daylighting and view objectives on which the façade was originally based. This is particularly true for 2‐dimensional systems, such as perforated metal screens. Because of this reason, the hypothesis of this research is that three‐dimensional systems can lead to better energy and IEQ performance outcomes compared with two‐dimensional geometry. For early phase design, overall performance of fenestration not only depends on parameters which include glare discomfort probability, useful daylight illuminance and solar radiation, but also refers to visual connection to the outdoors (view). Nevertheless, commonly existing researches did not examine view with other performance parameters simultaneously, but architects concern the outcome of view more than the effect of other parameters. Thus, the research examined the geometries in 2D and 3D dimension with four parameters including view, DGP, UDI and solar radiation. The perforated screen represented 2D base case and the louvers were 3D tested case. The measurements were simulated in Los Angeles and Seattle by Ladybug and Honeybee, and the louvers were from 0° to 75° in matte and specular materials. To develop the concept of design guide, the weighted score by linear transformation ranked the overall performance of those geometries, and the paired comparison was used for understanding potential sacrifices. As a result, when the opening was south‐facing, the louvers worked better than the perforated screen. In terms of design concept, for an architect, the best performance is the horizontal louver in Los Angele and Seattle except for considering both of view and glare in Seattle, and the exceptional louver is in 45 degrees. For an engineer, the best performance is the louver in 30 degrees in Los Angeles, but the louver in 45 degrees is better choice in Seattle. The potential sacrifice is glare issue based on the architecture choice, and part of illuminance maintenance might loss in terms of the engineering option. Based on the simulated measurement, the overall performance in three dimensions is better than it in two dimensions, and the results provide concept to develop the design guide in specific requirements to correspond with environmental conditions.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Multi-domain assessment of a kinetic facade: determining the control strategy of a kinetic façade using BIM based on energy performance, daylighting, and occupants’ preferences; Multi-domain asse...

PDF

The intelligent control strategy of kinetic façades for daylight and energy performance: evaluating the daylight effect of adaptive systems based on parametric workflow

PDF

Adaptive façade controls: A methodology based on occupant visual comfort preferences and cluster analysis

PDF

Performative shading design: parametric based measurement of shading system configuration effectiveness and trends

PDF

Exterior glare simulation: understanding solar convergence from concave facades using heat maps

PDF

Double skin façades performance: effects on daylight and visual comfort in office spaces

PDF

Comparing visual comfort metrics for fourteen spaces using simulation-based luminance mapping

PDF

Evaluation of daylighting circadian effects: Integrating non-visual effects of lighting in the evaluation of daylighting designs

PDF

Dynamic shading and glazing technologies: improve energy, visual, and thermal performance

PDF

Blind to light loss: evaluating light loss through commercial building facades as a contribution to urban light pollution

PDF

Effective light shelf and form finding: development of a light shelf design assistant tool using parametric methods

PDF

Environmentally responsive buildings: multi-objective optimization workflow for daylight and thermal quality

PDF

Absolute glare and relative glare factors: predicting and quantifying levels of interior glare and exterior glare caused by sunlight and daylight

PDF

Kinetic facades as environmental control systems: using kinetic facades to increase energy efficiency and building performance in office buildings

PDF

Pushing the solar envelope: determining solar envelope generating principles for sites with existing buildings

PDF

Tilted glazing: angle-dependence of direct solar heat gain and form-refining of complex facades

PDF

Development of circadian-effective toplighting strategies: using multiple daylighting performance goals for dementia care communities

Asset Metadata

Creator Chang, Hui Ling (author)

Core Title Evaluation and development of solar control screens: using daylight simulation to improve the performance of facade solar control screens

School School of Architecture

Degree Master of Science

Degree Program Building Science

Publication Date 04/20/2015

Defense Date 05/23/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag daylight glare probability,daylight harvesting,OAI-PMH Harvest,solar radiation,three‐dimensional geometry,two‐dimensional geometry,useful daylight illuminance,visual comfort

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Konis, Kyle (committee chair), Noble, Douglas (committee member), Schiler, Marc (committee member)

Creator Email huiling@usc.edu,viviolucky@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-553317

Unique identifier UC11297370

Identifier etd-ChangHuiLi-3336.pdf (filename),usctheses-c3-553317 (legacy record id)

Legacy Identifier etd-ChangHuiLi-3336.pdf

Dmrecord 553317

Document Type Thesis

Format application/pdf (imt)

Rights Chang, Hui Ling

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA