University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

DYNAMIC SHADING AND GLAZING TECHNOLOGIES:
Improve Energy, Visual, and Thermal Performance

by

Weixuan Lu

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

May 2022

Copyright 2022 Weixuan Lu

ii
ACKNOWLEDGEMENTS
I would like to express my deepest gratitude and appreciation to my committee chair, Professor
Kyle Konis, and my committee members, Professor Marc Schiler and Professor Douglas Noble.
They have been providing great help to my research and thesis from the beginning to the end.
What I have learned during this process will benefit me for the rest of my life.
I would like to extend my gratitude to Professor Karen Kensek for assisting me to finish the
thesis on time, and to faculty members and students for providing advice and comments to my
research.
I am also grateful to the Ladybug Tools Forum for providing technical suggestions and solutions
to numerous problems I have encountered during the research process.
Last but not least, I would like to offer special thanks to the MBS family for their mutual support
during the difficult pandemic years, and my own family for their encouragement overseas.
Committee:
Kyle Konis, AIA, Ph.D., Associate Professor, USC School of Architecture, kkonis@usc.edu
Marc E. Schiler, FASES, LC, Professor, USC School of Architecture, marcs@usc.edu
Douglas E. Noble, FAIA, Ph.D., Associate Professor, USC School of Architecture,
dnoble@usc.edu

iii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................ ii
LIST OF TABLES ......................................................................................................................... vi
LIST OF FIGURES ....................................................................................................................... ix
ABSTRACT ................................................................................................................................. xiii
CHAPTER ONE INTRODUCTION ...............................................................................................1
1.1 Problem .............................................................................................................................1
1.2 Approach ...........................................................................................................................2
1.3 Case Studies of Shading Systems......................................................................................4
1.3.1 Bullitt Center ..............................................................................................................4
1.3.2 Santa Monica City Services Building ........................................................................5
1.3.3 Entrada .......................................................................................................................6
1.4 Dynamic Glazing Techniques ...........................................................................................8
1.5 Combinatorial Dynamic Glazing and Shading Systems [Test Condition #6 & #7] ..........9
1.6 Significance of Testing Combinatorial Systems .............................................................10
1.7 Methodology Overview...................................................................................................11
1.8 Use of Software Programs...............................................................................................12
1.9 Summary .........................................................................................................................12
CHAPTER TWO LITERATURE REVIEW .................................................................................13
2.1 Forms of Kinetic Façades ................................................................................................13
2.2 Performance Goals of Energy .........................................................................................16
2.3 Simulation Tools of Electrochromic Glass .....................................................................17
2.4 Parameters of Electrochromic Glass ...............................................................................19
2.5 Visual Comfort and Thermal Comfort Metrics ...............................................................23
2.6 Full Scale Office Testbed Facilities ................................................................................27
2.7 Conclusions .....................................................................................................................27
CHAPTER THREE METHODOLOGY .......................................................................................29
3.1 Analyze Existing Shading Technologies.........................................................................30
3.2 Preliminary Shoebox Model Text [Experiment #1] ........................................................31
3.2.1 Select Software Programs ........................................................................................31
3.2.2 Model a Simple Room with Windows .....................................................................31
3.2.3 States Change During a Day ....................................................................................32
3.2.4 Simulate Visual, Energy, and Thermal Results .......................................................33
3.3 List of Test Conditions ....................................................................................................34
3.3.1 Clear Glazing Without Exterior Shading [Test Condition #1] ................................35
3.3.2 Kinetic Blinds [Test Conditions #2 & #3] ...............................................................35
3.3.3 Static Overhang [Test Condition #4] .......................................................................36
3.3.4 EC Glazing ...............................................................................................................37
3.3.5 Combined Systems of EC Glazing and Kinetic Blinds [Test Conditions #6 & #7] 37
3.3.6 Combined System of EC Glazing and Static Overhang ..........................................37

iv
3.4 Determining Performance Indicators ..............................................................................38
3.4.1 Visual Comfort.........................................................................................................38
3.4.2 Thermal Comfort .....................................................................................................40
3.4.3 Energy Use ...............................................................................................................41
3.4.4 Performance Scoring System ...................................................................................43
3.5 Model a Real Office Building [Experiment #2] ..............................................................44
3.5.1 The Entrada Project..................................................................................................45
3.6 Controllers of Dynamic Systems.....................................................................................47
3.7 Performance Goals ..........................................................................................................49
3.7.1 Energy ......................................................................................................................49
3.7.2 Visual Comfort.........................................................................................................49
3.7.3 Thermal Comfort .....................................................................................................50
3.7.4 View Quality ............................................................................................................50
3.8 Summary .........................................................................................................................51
CHAPTER FOUR RESULTS .......................................................................................................52
4.1 Experiment #1 - Shoebox Model ....................................................................................53
4.1.1 Daylighting ..............................................................................................................53
4.1.2 Glare .........................................................................................................................65
4.1.3 Peak Cooling Load ...................................................................................................74
4.1.4 EUI ...........................................................................................................................79
4.1.5 Adaptive Thermal Comfort ......................................................................................84
4.1.6 PMV Thermal Comfort ............................................................................................87
4.1.7 PPD ..........................................................................................................................93
4.2 Experiment #2 - The Entrada Project ............................................................................100
4.2.1 Daylighting ............................................................................................................100
4.2.2 Glare .......................................................................................................................112
4.2.3 Peak Cooling Load .................................................................................................122
4.2.4 EUI .........................................................................................................................127
4.2.5 Adaptive Thermal Comfort ....................................................................................131
4.2.6 PMV Thermal Comfort ..........................................................................................133
4.2.7 PPD ........................................................................................................................138
4.3 Conclusion .....................................................................................................................144
CHAPTER FIVE ANALYSIS .....................................................................................................145
5.1 Experiment #1 - The Shoebox Model ...........................................................................147
5.1.1 Daylighting ............................................................................................................147
5.1.2 Glare .......................................................................................................................149
5.1.3 Peak Cooling Loads ...............................................................................................153
5.1.4 EUI .........................................................................................................................154
5.1.5 PMV Thermal Comfort ..........................................................................................155
5.1.6 PPD ........................................................................................................................156
5.1.7 Performance Goals .................................................................................................157
5.1.8 Final Scores and Ranks ..........................................................................................160
5.2 Experiment #2 - The Entrada Project ............................................................................162
5.2.1 Daylighting ............................................................................................................162
5.2.2 Glare .......................................................................................................................164

v
5.2.3 Peak Cooling Loads ...............................................................................................168
5.2.4 EUI .........................................................................................................................168
5.2.5 PMV Thermal Comfort ..........................................................................................169
5.2.6 PPD ........................................................................................................................170
5.2.7 Performance Goals .................................................................................................171
5.2.8 Final Scores and Ranks ..........................................................................................174
CHAPTER SIX DISCUSSION & FUTURE WORK .................................................................177
6.1 Discussion .....................................................................................................................177
6.1.1 Key Findings ..........................................................................................................179
6.2 Future Work ..................................................................................................................186
6.2.1 Short-term Problems ..............................................................................................186
6.2.2 Long-term Goals ....................................................................................................189
6.3 Conclusion .....................................................................................................................192
REFERENCES ............................................................................................................................194

vi
LIST OF TABLES
Table 2-1 NREL Evaluation of ASHRAE 90.1-2007 (Long 2010) ............................................. 17
Table 2-2 Commercial Buildings Energy Consumption Survey (CBECS 2012) ......................... 17
Table 3-1 Information of the Shoebox Model .............................................................................. 34
Table 3-2 Test Conditions of the Shoebox Model ........................................................................ 35
Table 3-3 A Scoring Sample of All Test Conditions .................................................................... 44
Table 3-4 Information of Test Conditions of the Entrada Project ................................................ 47
Table 3-5 Information of the Entrada Project ............................................................................... 48
Table 4-1 Abbreviations of Test Conditions ................................................................................. 52
Table 4-2 States of Dynamic Systems on 9/21 ............................................................................. 53
Table 4-3 Daylighting Data of All Test Conditions...................................................................... 54
Table 4-4 Glare Situations of All Test Conditions ....................................................................... 65
Table 4-5 Peak Cooling Loads of All Test Conditions ................................................................. 75
Table 4-6 EUI Breakdowns of All Test Conditions...................................................................... 80
Table 4-7 EUI Breakdowns of Test Condition #1 ........................................................................ 80
Table 4-8 EUI Breakdowns of Test Condition #2 ........................................................................ 81
Table 4-9 EUI Breakdowns of Test Condition #3 ........................................................................ 81
Table 4-10 EUI Breakdowns of Test Condition #4 ...................................................................... 82
Table 4-11 EUI Breakdowns of Test Condition #5 ...................................................................... 82
Table 4-12 EUI Breakdowns of Test Condition #6 ...................................................................... 83
Table 4-13 EUI Breakdowns of Test Condition #7 ...................................................................... 83
Table 4-14 EUI Breakdowns of Test Condition #8 ...................................................................... 84
Table 4-15 Thermal Information of Test Condition #1 ................................................................ 86
Table 4-16 Thermal Information of Test Condition #4 ................................................................ 86
Table 4-17 Thermal Conditions of All Test Conditions ............................................................... 88
Table 4-18 Thermal Condition of Test Condition #1 ................................................................... 89
Table 4-19 Thermal Condition of Test Condition #2 ................................................................... 89
Table 4-20 Thermal Condition of Test Condition #3 ................................................................... 90
Table 4-21 Thermal Condition of Test Condition #4 ................................................................... 90
Table 4-22 Thermal Condition of Test Condition #5 ................................................................... 91
Table 4-23 Thermal Condition of Test Condition #6 ................................................................... 91
Table 4-24 Thermal Condition of Test Condition #7 ................................................................... 92
Table 4-25 Thermal Condition of Test Condition #8 ................................................................... 92
Table 4-26 PPD and Indoor Temperatures of All Test Conditions .............................................. 94
Table 4-27 PPD and Temperatures of Test Condition #1 ............................................................. 95
Table 4-28 PPD and Temperatures of Test Condition #2 ............................................................. 95
Table 4-29 PPD and Temperatures of Test Condition #3 ............................................................. 96
Table 4-30 PPD and Temperatures of Test Condition #4 ............................................................. 96
Table 4-31 PPD and Temperatures of Test Condition #5 ............................................................. 97
Table 4-32 PPD and Temperatures of Test Condition #6 ............................................................. 97
Table 4-33 PPD and Temperatures of Test Condition #7 ............................................................. 98
Table 4-34 PPD and Temperatures of Test Condition #8 ............................................................. 99
Table 4-35 States of Dynamic Systems of West and South Facades on 9/21 ............................. 101
Table 4-36 Daylighting Data of All Test Conditions.................................................................. 101
Table 4-37 Glare Situations of All Test Conditions ................................................................... 113

vii
Table 4-38 Peak Cooling Loads of All Test Conditions ............................................................. 122
Table 4-39 EUI Breakdowns of All Test Conditions.................................................................. 127
Table 4-40 EUI Breakdowns of Test Condition #1 .................................................................... 127
Table 4-41 EUI Breakdowns of Test Condition #2 .................................................................... 128
Table 4-42 EUI Breakdowns of Test Condition #3 .................................................................... 128
Table 4-43 EUI Breakdowns of Test Condition #4 .................................................................... 129
Table 4-44 EUI Breakdowns of Test Condition #5 .................................................................... 129
Table 4-45 EUI Breakdowns of Test Condition #6 .................................................................... 130
Table 4-46 EUI Breakdowns of Test Condition #7 .................................................................... 130
Table 4-47 EUI Breakdowns of Test Condition #8 .................................................................... 130
Table 4-48 Thermal Information of Test Condition #1 .............................................................. 132
Table 4-49 Thermal Information of Test Condition #4 .............................................................. 133
Table 4-50 PMV Thermal Conditions of All Test Conditions ................................................... 134
Table 4-51 Thermal Condition of Test Condition #1 ................................................................. 134
Table 4-52 Thermal Condition of Test Condition #2 ................................................................. 135
Table 4-53 Thermal Condition of Test Condition #3 ................................................................. 135
Table 4-54 Thermal Condition of Test Condition #4 ................................................................. 136
Table 4-55 Thermal Condition of Test Condition #5 ................................................................. 136
Table 4-56 Thermal Condition of Test Condition #6 ................................................................. 137
Table 4-57 Thermal Condition of Test Condition #7 ................................................................. 137
Table 4-58 Thermal Condition of Test Condition #8 ................................................................. 138
Table 4-59 PPD and Indoor Temperatures of All Test Conditions ............................................ 139
Table 4-60 PPD and Temperatures of Test Condition #1 ........................................................... 140
Table 4-61 PPD and Temperatures of Test Condition #2 ........................................................... 140
Table 4-62 PPD and Temperatures of Test Condition #3 ........................................................... 141
Table 4-63 PPD and Temperatures of Test Condition #4 ........................................................... 141
Table 4-64 PPD and Temperatures of Test Condition #5 ........................................................... 142
Table 4-65 PPD and Temperatures of Test Condition #6 ........................................................... 143
Table 4-66 PPD and Temperatures of Test Condition #7 ........................................................... 143
Table 4-67 PPD and Temperatures of Test Condition #8 ........................................................... 144
Table 5-1 Score and Rank Rules ................................................................................................. 146
Table 5-2 The Overall Scoring and Ranking Sample ................................................................. 146
Table 5-3 Sufficient Daylighting of Each Hour .......................................................................... 148
Table 5-4 Daylighting Score ....................................................................................................... 149
Table 5-5 Hourly DGP ................................................................................................................ 151
Table 5-6 Glare Score ................................................................................................................. 153
Table 5-7 Peak Cooling Score .................................................................................................... 154
Table 5-8 EUI Score ................................................................................................................... 155
Table 5-9 PMV Thermal Comfort Score .................................................................................... 156
Table 5-10 PPD Score ................................................................................................................. 157
Table 5-11 EUI Performance Goal ............................................................................................. 157
Table 5-12 sDA Performance Goal............................................................................................. 158
Table 5-13 sGA Performance Goal............................................................................................. 159
Table 5-14 Temperature Performance Goal ............................................................................... 159
Table 5-15 Dynamic States of the Shoebox Model .................................................................... 160
Table 5-16 View Performance Goal ........................................................................................... 160

viii
Table 5-17 Overall Performance Score of the Shoebox Model .................................................. 161
Table 5-18 Hourly Sufficient Daylighting .................................................................................. 163
Table 5-19 Daylighting Score ..................................................................................................... 164
Table 5-20 Hourly DGP .............................................................................................................. 166
Table 5-21 Glare Score ............................................................................................................... 167
Table 5-22 Peak Cooling Score .................................................................................................. 168
Table 5-23 EUI Score ................................................................................................................. 169
Table 5-24 Thermal Score .......................................................................................................... 170
Table 5-25 PPD Score ................................................................................................................. 171
Table 5-26 EUI Performance Goal ............................................................................................. 172
Table 5-27 sDA Performance Goal............................................................................................. 172
Table 5-28 sGA Performance Goal............................................................................................. 173
Table 5-29 Temperature Performance Goal ............................................................................... 173
Table 5-30 Dynamic States of the Entrada Project ..................................................................... 174
Table 5-31 View Performance Goal ........................................................................................... 174
Table 5-32 Overall Performance Score of the Entrada Project ................................................... 175
Table 6-1 Default Heating and Cooling Setpoints for the Large Office Buildings .................... 184
Table 6-2 Adaptive Thermal Comfort of Experiments #1 and #2 .............................................. 188

ix
LIST OF FIGURES
Figure 1-1 Observations of Interior Shades on the North Facade .................................................. 2
Figure 1-2 Observations of Interior Shades on the West Façade with Exterior Shades ................. 2
Figure 1-3 Observations of Interior Shades on the South Facade .................................................. 2
Figure 1-4 Observations of Interior Shades on the East Facade ..................................................... 2
Figure 1-5 Temperatures of Los Angeles ....................................................................................... 3
Figure 1-6 Bullitt Center (Bullitt Center 2013) .............................................................................. 5
Figure 1-7 Santa Monica City Services Building (Herd 2021) ....................................................... 6
Figure 1-8 Entrada (Vincent 2018) ................................................................................................. 7
Figure 1-9 EC Windows in Clear and Tinted States (Lee 2016) .................................................. 10
Figure 2-1 Foldable Shade, Horizontal Blinds, Vertical Blinds, and Overhang ........................... 14
Figure 2-2 Shading Geometries Abstracted from Leaves Surface and Insects Hindwings .......... 14
Figure 2-3 South Façade of the 911 Federal Building (Lee 2016) ............................................... 20
Figure 2-4 Four Tinted Levels of EC (Lee 2016) ......................................................................... 20
Figure 2-5 South Façade of the Moss Federal Building (Fernandes 2021) .................................. 21
Figure 2-6 Subzones of Windows (Fernandes 2021) .................................................................... 21
Figure 3-1 Methodology Overview .............................................................................................. 30
Figure 3-2 Shoebox Model ........................................................................................................... 32
Figure 3-3 Test Condition 1 .......................................................................................................... 36
Figure 3-4 Test Condition 2 .......................................................................................................... 36
Figure 3-5 Test Condition 3 .......................................................................................................... 36
Figure 3-6 Test Condition 4 .......................................................................................................... 36
Figure 3-7 Test Condition 5 .......................................................................................................... 38
Figure 3-8 Test Condition 6 .......................................................................................................... 38
Figure 3-9 Test Condition 7 .......................................................................................................... 38
Figure 3-10 Test Condition 8 ........................................................................................................ 38
Figure 3-11 The Full Site of Entrada with Scale .......................................................................... 46
Figure 3-12 The Third Floor Plan and Model of Entrada (LPC West 2021) ................................ 46
Figure 3-13 Southwest and Northeast Isometrics of the Third Floor Corner of Entrada ............. 46
Figure 4-1 Shoebox Model ........................................................................................................... 53
Figure 4-2 Scale of Illuminance .................................................................................................... 54
Figure 4-3 Momentary Illuminance of Test Condition #1 ............................................................ 55
Figure 4-4 Momentary Illuminance of Test Condition #2 ............................................................ 57
Figure 4-5 Momentary Illuminance of Test Condition #3 ............................................................ 58
Figure 4-6 Momentary Illuminance of Test Condition #4 ............................................................ 59
Figure 4-7 Momentary Illuminance of Test Condition #5 ............................................................ 61
Figure 4-8 Momentary Illuminance of Test Condition #6 ............................................................ 62
Figure 4-9 Momentary Illuminance of Test Condition #7 ............................................................ 63
Figure 4-10 Momentary Illuminance of Test Condition #8 .......................................................... 64
Figure 4-11 Fisheye Scene ............................................................................................................ 65
Figure 4-12 Orientation................................................................................................................. 65
Figure 4-13 Scale of Sky Luminance............................................................................................ 65
Figure 4-14 Sky Luminance and DGP of Test Condition #1 ........................................................ 67
Figure 4-15 Sky Luminance and DGP of Test Condition #2 ........................................................ 68
Figure 4-16 Sky Luminance and DGP of Test Condition #3 ........................................................ 69

x
Figure 4-17 Sky Luminance and DGP of Test Condition #4 ........................................................ 70
Figure 4-18 Sky Luminance and DGP of Test Condition #5 ........................................................ 71
Figure 4-19 Sky Luminance and DGP of Test Condition #6 ........................................................ 72
Figure 4-20 Sky Luminance and DGP of Test Condition #7 ........................................................ 73
Figure 4-21 Sky Luminance and DGP of Test Condition #8 ........................................................ 74
Figure 4-22 Peak Cooling Load of Test Condition #1 .................................................................. 75
Figure 4-23 Peak Cooling Load of Test Condition #2 .................................................................. 76
Figure 4-24 Peak Cooling Load of Test Condition #3 .................................................................. 76
Figure 4-25 Peak Cooling Load of Test Condition #4 .................................................................. 77
Figure 4-26 Peak Cooling Load of Test Condition #5 .................................................................. 77
Figure 4-27 Peak Cooling Load of Test Condition #6 .................................................................. 78
Figure 4-28 Peak Cooling Load of Test Condition #7 .................................................................. 78
Figure 4-29 Peak Cooling Load of Test Condition #8 .................................................................. 79
Figure 4-30 Neutral Temperatures of the Year ............................................................................. 85
Figure 4-31 Adaptive Thermal Comfort of Test Condition #1 ..................................................... 85
Figure 4-32 Deviation from Neutral Temperatures of Test Condition #1 .................................... 86
Figure 4-33 Adaptive Thermal Comfort of Test Condition #4 ..................................................... 86
Figure 4-34 Deviation from Neutral Temperatures of Test Condition #4 .................................... 87
Figure 4-35 Thermal Condition of Test Condition #1 .................................................................. 88
Figure 4-36 Thermal Condition of Test Condition #2 .................................................................. 89
Figure 4-37 Thermal Condition of Test Condition #3 .................................................................. 89
Figure 4-38 Thermal Condition of Test Condition #4 .................................................................. 90
Figure 4-39 Thermal Condition of Test Condition #5 .................................................................. 90
Figure 4-40 Thermal Condition of Test Condition #6 .................................................................. 91
Figure 4-41 Thermal Condition of Test Condition #7 .................................................................. 92
Figure 4-42 Thermal Condition of Test Condition #8 .................................................................. 92
Figure 4-43 Scale of PPD ............................................................................................................. 94
Figure 4-44 PPD of Test Condition #1 ......................................................................................... 94
Figure 4-45 PPD of Test Condition #2 ......................................................................................... 95
Figure 4-46 PPD of Test Condition #3 ......................................................................................... 96
Figure 4-47 PPD of Test Condition #4 ......................................................................................... 96
Figure 4-48 PPD of Test Condition #5 ......................................................................................... 97
Figure 4-49 PPD of Test Condition #6 ......................................................................................... 97
Figure 4-50 PPD of Test Condition #7 ......................................................................................... 98
Figure 4-51 PPD of Test Condition #8 ......................................................................................... 98
Figure 4-52 Southwest and Northeast Isometric Views of Entrada Office Corner .................... 100
Figure 4-53 Scale Momentary Illuminance ................................................................................ 101
Figure 4-54 Momentary Illuminance of Test Condition #1 ........................................................ 102
Figure 4-55 Momentary Illuminance of Test Condition #2 ........................................................ 104
Figure 4-56 Momentary Illuminance of Test Condition #3 ........................................................ 105
Figure 4-57 Momentary Illuminance of Test Condition #4 ........................................................ 106
Figure 4-58 Momentary Illuminance of Test Condition #5 ........................................................ 107
Figure 4-59 Momentary Illuminance of Test Condition #6 ........................................................ 109
Figure 4-60 Momentary Illuminance of Test Condition #7 ........................................................ 110
Figure 4-61 Momentary Illuminance of Test Condition #8 ........................................................ 111
Figure 4-62 Orientation............................................................................................................... 112

xi
Figure 4-63 Scale of Sky Luminance.......................................................................................... 113
Figure 4-64 Sky Luminance and DGP of Test Condition #1 ...................................................... 114
Figure 4-65 Sky Luminance and DGP of Test Condition #2 ...................................................... 115
Figure 4-66 Sky Luminance and DGP of Test Condition #3 ...................................................... 116
Figure 4-67 Sky Luminance and DGP of Test Condition #4 ...................................................... 117
Figure 4-68 Sky Luminance and DGP of Test Condition #5 ...................................................... 118
Figure 4-69 Sky Luminance and DGP of Test Condition #6 ...................................................... 119
Figure 4-70 Sky Luminance and DGP of Test Condition #7 ...................................................... 120
Figure 4-71 Sky Luminance and DGP of Test Condition #8 ...................................................... 121
Figure 4-72 Peak Cooling Load of Test Condition #1 ................................................................ 122
Figure 4-73 Peak Cooling Load of Test Condition #2 ................................................................ 123
Figure 4-74 Peak Cooling Load of Test Condition #3 ................................................................ 124
Figure 4-75 Peak Cooling Load of Test Condition #4 ................................................................ 124
Figure 4-76 Peak Cooling Load of Test Condition #5 ................................................................ 125
Figure 4-77 Peak Cooling Load of Test Condition #6 ................................................................ 125
Figure 4-78 Peak Cooling Load of Test Condition #7 ................................................................ 126
Figure 4-79 Peak Cooling Load of Test Condition #8 ................................................................ 126
Figure 4-80 Neutral Temperatures of the Year ........................................................................... 131
Figure 4-81 Adaptive Thermal Comfort of Test Condition #1 ................................................... 132
Figure 4-82 Deviation from Neutral Temperatures of Test Condition #1 .................................. 132
Figure 4-83 Adaptive Thermal Comfort of Test Condition #1 ................................................... 132
Figure 4-84 Deviation from Neutral Temperatures of Test Condition #4 .................................. 133
Figure 4-85 Thermal Condition of Test Condition #1 ................................................................ 134
Figure 4-86 Thermal Condition of Test Condition #2 ................................................................ 135
Figure 4-87 Thermal Condition of Test Condition #3 ................................................................ 135
Figure 4-88 Thermal Condition of Test Condition #4 ................................................................ 136
Figure 4-89 Thermal Condition of Test Condition #5 ................................................................ 136
Figure 4-90 Thermal Condition of Test Condition #6 ................................................................ 137
Figure 4-91 Thermal Condition of Test Condition #7 ................................................................ 137
Figure 4-92 Thermal Condition of Test Condition #8 ................................................................ 138
Figure 4-94 Scale of PPD ........................................................................................................... 139
Figure 4-94 PPD of Test Condition #1 ....................................................................................... 139
Figure 4-95 PPD of Test Condition #2 ....................................................................................... 140
Figure 4-96 PPD of Test Condition #3 ....................................................................................... 141
Figure 4-97 PPD of Test Condition #4 ....................................................................................... 141
Figure 4-98 PPD of Test Condition #5 ....................................................................................... 142
Figure 4-99 PPD of Test Condition #6 ....................................................................................... 142
Figure 4-100 PPD of Test Condition #7 ..................................................................................... 143
Figure 4-101 PPD of Test Condition #8 ..................................................................................... 144
Figure 5-1 Line Chart of Hourly Sufficient Daylighting ............................................................ 147
Figure 5-2 Bar Chart of Average Sufficient Daylighting ........................................................... 148
Figure 5-3 Line Chart of Hourly DGP ........................................................................................ 150
Figure 5-4 3D Line Chart of Hourly DGP .................................................................................. 151
Figure 5-5 Bar Chart of Glare Situations .................................................................................... 152
Figure 5-6 Bar Chart of Peak Cooling Loads ............................................................................. 153
Figure 5-7 Highest and Lowest Peak Cooling Loads ................................................................. 153

xii
Figure 5-8 Bar Chart of EUI Breakdowns .................................................................................. 154
Figure 5-9 Bar Chart of PMV Thermal Comfort ........................................................................ 155
Figure 5-10 Bar Chart of PPD .................................................................................................... 156
Figure 5-11 Line Chart of Hourly Sufficient Daylighting .......................................................... 162
Figure 5-12 Bar Chart of Average Sufficient Daylighting ......................................................... 163
Figure 5-13 Line Chart of Hourly DGP ...................................................................................... 165
Figure 5-14 3D Line Chart of Hourly DGP ................................................................................ 166
Figure 5-15 Bar Chart of Glare Situations .................................................................................. 167
Figure 5-16 Bar Chart of Peak Cooling Loads ........................................................................... 168
Figure 5-17 Highest and Lowest Peak Cooling Loads ............................................................... 168
Figure 5-18 Bar Chart of EUI Breakdowns ................................................................................ 169
Figure 5-19 Bar Chart of PMV Thermal Comfort ...................................................................... 170
Figure 5-20 Bar Chart of Highest PPD % ................................................................................... 171

xiii
ABSTRACT
Conventional static glazing sometimes has poor performance in energy, visual, and thermal
aspects. In this thesis, a series of simulations of an office building were done to compare the
performance of conventional static glazing, exterior static and kinetic shades, dynamic glazing,
and dynamic glazing working together with static or kinetic shades as a combinatorial system.
Energy uses include annual Energy Use Intensity (EUI) and peak cooling loads of the hottest day
of the year. Visual performance includes maximizing daylight and minimizing glare. Thermal
performance includes Predicted Mean Vote (PMV) thermal comfort that was based on
occupants’ sensation of thermal conditions and Predicted Percentage of Dissatisfied (PPD)
which indicated the levels of thermal discomfort. A scoring system was developed to evaluate
the overall performance. The prediction was that the combination of dynamic glazing and kinetic
shade would have the best comprehensive performance among visual, energy, and thermal
comfort.
Key Words: Energy, Daylight, Glare, Thermal Comfort, Electrochromic Glass, Kinetic Shades,
Grasshopper, Ladybug Tools
Hypothesis: A window shading system combining dynamic glass and exterior kinetic shades has
superior performance with regard to energy use, glare, and thermal comfort than the following
systems used in isolation: static glazing (VLT = 0.64, SHGC = 0.42), static shades, dynamic
glazing, and kinetic shades.

1
CHAPTER ONE INTRODUCTION
1.1 Problem
Facades with large window areas provide sufficient daylight to compensate for artificial lighting
but may cause more energy consumption. A high window-to-wall ratio (WWR) brings more
daylight to the interior, but it also can cause more glare and capture excessive solar heat gain.
Daylight access, solar heat gains, energy consumption, and indoor environmental quality are
difficult issues for designers to balance. Static shading systems help occupants to find trade-offs
between comfort and discomfort, but there is often a reduction of flexibility to control the
shading. More advanced glazing and shading technologies can help address the issues. Dynamic
§shading systems could be adaptive to the changing climates or occupants’ comfort. There are
types of dynamic glass that turn tinted when too much heat or light is going through the glass.
However, advanced technology cannot cover all the gaps. The kinetic shading or dynamic glass
applied to a building usually aims to resolve or emphasize only one problem while overlooking
others. The complexities of climate changes and human needs make it difficult to find the most
appropriate balance. While considering the visual comfort of daylight and glare, energy
consumption, and indoor thermal comfort, a package of multiple shading techniques should be
addressed. A solar shading system with a combination of dynamic glazing and kinetic shading
may help buildings meet performance goals of energy use, visual comfort, and thermal comfort.
The images below (FIG. 1-1, 1-2, 1-3, 1-4) show the results from an observational study of
several buildings located in Los Angeles around the fall equinox. Images of these buildings at the
University of Southern California (USC) campus were taken at different times of the day.
Facades facing different orientations with large window areas have most of the interior shades

2
deployed all day. Even the example with exterior shading on the façade still has most of the
shades down. Those windows do not provide sufficient daylight but still gain solar heat. The
transparency goal of obtaining daylight is not achieved in the design.
8 am 12 pm 4 pm 6 pm

Figure 1-1 Observations of Interior Shades on the North Facade

Figure 1-2 Observations of Interior Shades on the West Façade with Exterior Shades

Figure 1-3 Observations of Interior Shades on the South Facade

Figure 1-4 Observations of Interior Shades on the East Facade
1.2 Approach
Commercial buildings tend to have larger amounts of energy use and higher occupancy densities
than other building types (Godoy-Shimizu 2018). New construction of large office buildings

3
keeps taking place globally, and buildings with large window areas account for a large
proportion (Love 2019). Reducing energy consumption contributes to the environment by
reducing fossil fuels used and improving occupants’ comfort. The American Society of Heating,
Refrigerating and Air-Conditioning Engineers (ASHRAE) 90.1 provides different benchmarks of
Energy Use Intensity (EUI) for different climate zones, which means total energy consumption
by a building of a year. For shading systems, cold climate zones need more solar heat while hot
climate zones need less. Mixed climates have opposite conditions in summers and winters. A
package of dynamic shading and glazing technologies was predicted to help buildings in
different climates to achieve performance goals more easily and provide visual and thermal
comfort at the same time. This research tested the performance of office buildings in Los
Angeles, which has a dry warm climate (climate Zone 3B according to ASHRAE). The image
shows the annual temperatures of Los Angeles (FIG. 1-5). The combination of strategies should
help the buildings to achieve the following standards: less than 5% of the time having Daylight
Glare Probability (DGP) larger than 40% (sGA40%5%), and temperatures ranging from 67 to 82 °F
(ASHRAE Standard 55-2017) when the heating, ventilation, and air conditioning (HVAC)
systems are used. However, they still could not help the buildings to achieve energy use 80%
lower than the regional average (Architecture 2030 Challenge), and Spatial Daylight Autonomy
(sDA) of at least 55% (WELL Building Standard v1).

Figure 1-5 Temperatures of Los Angeles

4
1.3 Case Studies of Shading Systems
Strategies of static and kinetic shading have been continuously applied to different buildings for
years since they have been invented. There is plenty of available data that could evaluate whether
different kinds of shading technologies have been improving building performance and
occupants’ comfort.
1.3.1 Bullitt Center
The Bullitt Center (FIG. 1-6) in Seattle is known as one of the greenest commercial buildings in
the world (Bullitt Center 2013). Its Energy Use Intensity (EUI) is only 16 kBtu/ft
2
(50 kWh/m
2
).
The current Seattle Energy Code’s EUI requirement for an office building is 51 kBtu/ft
2
(161
kWh/m
2
). While a typical office building in Seattle usually has 72 kBtu/ft
2
(227 kWh/m
2
) of
EUI. The shading system of the curtain wall facade is an essential factor for achieving a low
EUI. The Bullitt Center has an adaptive curtainwall façade. The apertures have operable interior
rolling shades and exterior louver blinds to shield glare and unwanted solar heat gain. The
interior blinds are used to prevent glare while the exterior shades are used to block the heat. The
windows are designed not to influence the exterior shades when open. They are also automated
to improve passive ventilation. Occupants could only open or close the windows within a certain
range of temperatures. Once the outdoor air has reached an excessively hot or cold temperature,
the façade system automatically switches to the most energy-efficient mode to maintain the
interior environmental quality. Both windows and blinds operate themselves to adapt to the
changing climate conditions (Smith-Gardiner 2012). Nevertheless, located in a rainy city, the
Bullitt Center has a large extended roof of integrated PV panels to generate sufficient solar
energy. The roof itself creates shade on facade areas of the upper floors. This feature may not be
applicable to many office buildings in dense urban areas. The three separate control systems of

5
the façade are also complex. For large commercial buildings, a dynamic shading system with one
control system is more convenient for occupants, but it should better adapt to the climates and
occupants’ comfort.

Figure 1-6 Bullitt Center (Bullitt Center 2013)
1.3.2 Santa Monica City Services Building
The Santa Monica City Services Building (FIG. 1-7) is a commercial building that meets the
Living Building Challenge criteria, which is currently the most rigorous criteria for
sustainability. The building has a complicated glass curtain wall system. The curtain wall panels
have variable densities of frit patterns sandwiched between the two glazing panes to reduce solar
heat gain. There are operable windows and passive shades as the image shows. The façade
system uses passive techniques to maximize daylighting and natural ventilation and reduce
energy use. The manually controlled facade system improves both heating and cooling
performance while minimizing operation and maintenance costs. The building has a low EUI

6
value target at 21 kBtu/ft
2
(66 kWh/m
2
) per year (Herd 2021). However, the downside of a
manually operated system is that its benefit might be less frequent than expected because
occupants tend not to manually operate the shading systems very often (O'Brien 2013). This
might cause additional active cooling and heating loads, use of artificial lighting, and energy
consumption. The curtain wall with patterns sandwiched between glass panes also reduces the
available daylight compared to glazing without patterns. If the façade system is adaptive to
climate changes and occupant’s needs, the actual building performance may align more closely
with the design intent.

Figure 1-7 Santa Monica City Services Building (Herd 2021)
1.3.3 Entrada
The Entrada project (FIG. 1-8), designed by Gensler, is a large mix-used office building in
Culver City near Los Angeles. The structure has a curtain wall façade system with no separate,

7
exterior shading device. The design of numerous balconies provides shaded areas and
connections between interior and exterior space (Vincent 2018). However, the glass envelope
likely causes significant amounts of energy consumption, especially when it is located in south
California where solar heat is abundant. Interior shading could only shield the sunlight visually,
but the solar heat still enters the building and increase the indoor temperature. Then an increase
in cooling loads is caused. While no data on the building’s performance is available, this glass
building has the potential to improve its energy conservation.

Figure 1-8 Entrada (Vincent 2018)
Based on an assessment of the case studies, current shading technologies have opportunities to
be improved. To develop a more effective shading technology, one of the most important factors
is the implementation of automation and adaptivity to changing environments so that occupants
could be freed from the responsibility of controlling shades. The daily and seasonal variations of

8
weather conditions are the challenging part because the dynamic shading system has to respond
to all the changes to be functional.
1.4 Dynamic Glazing Techniques
Dynamic glass and kinetic façades have already been developed and applied to a large number of
buildings. They are controlled by sensors and could be overridden by occupants. When the
sensor detects the temperature, solar irradiance, or illuminance is above a certain level, the glass
becomes tinted, and the shades are deployed. Although they have a lot of environmental benefits,
they are generally too expensive to be in widespread use. Electrochromic (EC) glazing uses
electric currents between two glass panes to change tinted levels (Williams 2019). Manufacturers
claim EC glazing could save 20% energy from active cooling and artificial lighting, and 8%
energy from the overall building (Woodford 2011). Switching transparency levels consumes very
little electricity; however, manufacturing EC glazing is expensive and EC glazing has a shorter
life span compared to regular glazing materials (Woodford 2011). The question as to whether EC
glazing is worth installing to save energy and increase comfort gets various responses from
building owners.
The experiment presented in this thesis tested the performance of dynamic techniques including
EC glazing and kinetic shades. The solar heat gain coefficient (SHGC), which indicates the
fraction of solar radiance that goes through the glass (Sbar 2012), and visible light transmittance
(VLT) which means the amount of visible spectrum of light that passes through the glass (Sbar
2012) would be varied for the EC glazing. EC glazing has been tested to effectively reduce
SHGC. For instance, the clear state of EC glazing has a VLT of 60% and SHGC of 0.47. The
fully tinted state of EC glazing has a VLT of 2% and SHGC of 0.09 (Sbar 2012). However, solid

9
exterior shades are still more effective to shield solar heat. SHGC values are largely influenced
by the transmittance of glass. Also, in order to achieve the goal of 80% energy use less reduction
relative to the regional average, the needs for solar heat would be different in hot, mixed, and
cold climate zones. The dynamic glass alone could potentially shield sufficient unwanted solar
heat for buildings in cold climates. Those buildings require double glazing and thermal breaks to
minimize conductive heat gains or losses, and they also need to maximize daylighting and solar
heat gain. The EC glass would only be used to solve glare issues in this situation. For mixed
climates, hot and cold seasons would have different conditions. External shades would be
necessary for summer to cut solar heat gain, but not useful in winter when daylighting and solar
heating are needed. The external shades might be very important to the buildings located in hot
climates because although EC glazing can reduce transmissivity by around 90% when it is fully
tinted, much of the heat is absorbed by the glass and reradiates to the interior. On the other hand,
exterior shades can completely block the sunlight before it enters through the window.
1.5 Combinatorial Dynamic Glazing and Shading Systems [Test
Condition #6 & #7]
This research compared different types of shading strategies, including static and dynamic
glazing, and static and kinetic exterior shades. Two of the test conditions were combinatorial
systems of dynamic glazing and kinetic shading. The performance regarding energy, daylight,
glare and thermal comfort was simulated by Grasshopper software. In reality, electrochromic,
photochromic, and thermochromic glazing types have huge differences in the manufacturing
process. Electrochromic glass has electrical currents between two layers of glass to determine its
color and transmissivity (FIG.1-9), thermochromic glazing adjusts its transparency due to
temperature changes, and photochromic glass changes its tinted levels based on the levels of

10
ultraviolet (UV) radiation (Williams 2019). For this experiment, types of dynamic glazing did
not matter because the external stimuli could be easily changed by software programs, so EC
glazing was assumed to be the dynamic glazing used.

Figure 1-9 EC Windows in Clear and Tinted States (Lee 2016)
Different external stimuli could be used as controllers to switch tinted levels of glazing and
lower the kinetic shades. They would lead to different levels of energy efficiency. The
controllers of EC glazing and kinetic shades could be sun movements, temperatures, illuminance,
solar irradiance, or glare. However, they were not all tested in the experiment. The literature
review indicates which controller is the most commonly used for EC glazing in practice.
1.6 Significance of Testing Combinatorial Systems
The EC glazing itself has four stages: Clear, Light, Heavy, and Full. Deploying the kinetic
exterior shades would be the fifth stage of the combinatorial system while the glazing stays fully
tinted. The exterior shades could block the direct solar radiation. In the meantime, the kinetic
shading would not block daylight through the glazing when it was not activated.

11
It could be very expensive to realize a combination of EC glazing and kinetic exterior shading.
The kinetic shades or EC glass are each separately fairly expensive. However, even if the system
does not seem financially practical currently, it is important to test whether it has superior
performance or not. If the combination of EC glazing and kinetic exterior shading could reduce
energy loads and artificial lighting needs, and guarantee visual and thermal comfort at the same
time, it could be an effective shading option for future buildings. Technology continues to make
progress. In the near future, there will likely be less expensive or more accessible ways to
achieve energy reduction.
1.7 Methodology Overview
The complete methodology will be described in detail in Chapter 3. The first step is to test a
shoebox model with a 30% Window-to-Wall Ratio (WWR) because it is an efficient WWR
regarding daylight (Sayadi 2021). The weather data of Los Angeles was used. The next step is to
use the simulation tools to calculate the energy consumption, daylighting, glare, and thermal
levels of different test conditions. Then workflows were developed for each of the test
conditions, including static glazing, static glazing with static overhang, static glazing with
exterior kinetic shades, EC glazing, EC glazing with static overhang, and EC glazing with kinetic
shades. Two types of kinetic blinds were tested, so there were eight test conditions in total. The
performance results include average illuminance, Daylight Glare Probability (DGP), peak
cooling loads, Energy Use Intensity (EUI), Predicted Mean Vote (PMV) thermal comfort, and
Predicted Percentage Dissatisfied (PPD) thermal conditions. The developed workflows were then
applied to a real office building model. Finally, the results of different test conditions of the real
office model were compared and evaluated by a scoring system.

12
1.8 Use of Software Programs
Several simulation programs have been tested for the experiment. EnergyPlus and Radiance are
simulation engines to simulate thermal, energy, radiance, and daylight performance (Aspen
2018). Their calculating methods are used in different software for visualization. Grasshopper is
a plugin of model software Rhino. It has various plugins that could do the simulation of energy,
visual and thermal performance using EnergyPlus and Radiance. Multiple Grasshopper plugins
have been tested for the research, including DIVA, Honeybee Plus, and Ladybug Tools. The
simple shoebox model and simple daylight analyses have been simulated to test the plugins. The
Ladybug Tools is the most effective analytical software because it has the option to simulate
dynamic glazing and kinetic shades and could assign different VLT and SHGC to glass.
1.9 Summary
A combinatorial system of EC glazing and kinetic shading is predicted to have superior
performance of visual, energy, and thermal performance compared with other shading and
glazing techniques. However, the form of kinetic shading, the external stimuli to control the
dynamic system, and the control algorithms of VLT and SHGC of EC glazing will be further
explored in the literature review.

13
CHAPTER TWO LITERATURE REVIEW
Dynamic shading and glazing technologies have been invented to adapt to climates and
occupants’ comfort. However, due to the complexity of sky conditions, sun movements, and
human needs for visual and thermal comfort, there is always the potential to improve. Many
scholars have been researching methods to evaluate the efficiency of shading technologies. This
chapter describes a review of the literature of what has already been studied in the area of
dynamic shading and glazing technologies. Those studies guided this research to possible
improvements that could be made for the proposed test conditions and provided the research with
information including the control algorithm. This chapter also contains references that the
experiments of this thesis are based on.
2.1 Forms of Kinetic Façades
The kinetic façade has been widely used in building designs. The dynamic features of kinetic
facades have made them well-known for achieving high performance (Lee, Dong-Seok 2016).
While kinetic shading alone can reduce energy consumption so as to reduce fossil fuels used
(Lee, Dong-Seok 2016), it is assumed to work well with EC glazing as a combinatorial system.
The experiment also assessed energy efficiency, daylighting, glare, and thermal performance of
the combinatorial dynamic system as one of the test conditions.
Kinetic façades have various forms. Among the many forms are solid foldable panels, horizontal
louvers that could be raised, and vertical louver blinds that could change the angles of slats (Fig.
2-1). Significant previous research has investigated kinetic facades’ forms. Software programs
have been used to test the forms that are the best for building performance. Morteza Hosseini has
researched the important factors to improve the visual and thermal comfort of kinetic facades

14
(Hosseini 2019). He hypothesized that the geometries of façade panels are important factors to
adapt daylight for different climates. Generative-parametric and form-finding designs were used
to improve visual and thermal comfort. The design process also reflected the application of solar
radiation, wind resources, and dynamic daylight adaption (Hosseini 2019). Unusual forms like
structures of leaves and flowers’ surfaces and insects’ wings (FIG. 2-2) have been studied. The
results showed that for shading devices, semi-open spaces and changeable three-dimensional
(3D) shapes in façades could improve daylight and thermal performance the most (Hosseini
2019). In the experiment of this thesis, the simulation tools cannot support an excessively
complicated form of kinetic façade. To have forms of semi-open spaces and changeable 3D
shapes, louver blinds with different angles and slat widths are more suitable than foldable solid
panels or roller shades for visual and thermal comfort. Even if they were kinetic, zero amount of
daylight will be available with the deployment of shades. Therefore, fully covered roller shades
were not included in the test conditions.

Figure 2-1 Foldable Shade, Horizontal Blinds, Vertical Blinds, and Overhang

Figure 2-2 Shading Geometries Abstracted from Leaves Surface and Insects Hindwings

15
Energy conservation of kinetic facades is equally important to visual and thermal comfort. It was
demonstrated that kinetic shading had the potential to reduce 20% to 30% of the energy
consumption of commercial buildings (Hosseini 2019). Ryan Hansanuwat has proposed an
energy-efficient design of kinetic facades. He tested overhangs, foldable shades, vertical louver
shades, and horizontal louver shades based on simulation results of solar thermal, daylighting,
ventilation, and energy generation. Hansanuwat’s research indicated that the best solution as an
environmental mediator was vertical louver shades that also could be used as an overhang when
the slats were shut and the entire shading panel was rotated to horizontal. He has also built a
physical prototype façade panel (Hansanuwat 2010). Vertical louver blinds are more effective on
the east and west facades because of the sun angles. The curtain wall office buildings usually
have large window areas on all sides of the building, so the standard test model of the experiment
also had large glazing areas on all sides of the facades. Therefore, horizontal louver blinds were
tested as one of the shading options in the experiment. The exterior louvers could be lifted to
provide sufficient daylight when the solar heat is not disturbing. However, the simulation tools
could not automatically change slats’ tilt angles. Different angles and slats’ widths were tested
separately in the experiment. One test condition was the kinetic miniature blinds, and the other
condition had a wider slat, a larger spacing, and a different angle. Additionally, the frame of
louver blinds lifted as an overhang is too complicated for the simulation’s automation. Since the
overhang was tested to be effective for thermal comfort, daylight performance, and energy
conservation in those studies, the static horizontal overhang was tested and compared in the
experiment as another shading option.

16
2.2 Performance Goals of Energy
Multiple performance goals could be used to evaluate and calculate the efficiency of shading and
glazing strategies. Research of dynamic foldable overhang shading systems based on multiple
domains has been done by Kunyu Luo (Luo 2018). The adaptive shade was controlled by several
factors so as to provide all kinds of indoor comforts to occupants. Luo has used Revit Dynamo to
simulate important parameters like energy, solar heat gain, daylighting, and occupants’
behaviors. The solar heat gain and hourly indoor illuminance were calculated by the ASHRAE
handbook and DOE-2 Daylighting (Luo 2018). Benchmarking was important to evaluate the
progress that could be made by new shading technologies. When multiple factors were required
to be considered, multiple standards should be used as baselines or calculation standards. For
energy use of different building types from different climate zones, there was data from National
Renewable Energy Laboratory (NREL) Evaluation (Table 2-1) that evaluated data from
ASHRAE 90.1 and 189.1. Also, Commercial Buildings Energy Consumption Survey (CBECS)
(Table 2-2) collected information of commercial buildings in different regions in the U.S. For
ASHRAE 90.1, the EUI target of 3B climate zones in California was 95 kWh/m
2
, and for
CBECS, the EUI value for hot and dry climate zones was 185 kWh/m
2
. A further step was to use
the EUI data as baselines to see whether the test conditions could help the buildings to achieve
80% EUI lower than the baseline value to accomplish the Architecture 2030 Challenge. Newly
built or renovated buildings having energy use 70% less than the regional average of the same
building type accomplished the Architecture 2030 Challenge. The goal for 2020 was 80%
reduction and for 2025 was 90% (Architecture 2030). ASHRAE 90.1 provided the benchmark
EUI of commercial buildings, 189.1 provided the benchmark EUI of high-performance green
buildings, while CBECS provided the baseline EUI. Therefore, 185 kWh/m
2
was used as the

17
baseline EUI in the experiments, and the value with an 80% reduction was 37 kWh/m
2
. The test
conditions would accomplish the Architecture 2030 Challenge if they had EUI lower than 37
kWh/m
2
.
Table 2-1 NREL Evaluation of ASHRAE 90.1-2007 (Long 2010)
Building Type /
Energy Use
Intensity
(kWh/m
2
)
Climate Zone
1A 2A 2B 3A 3B:
CA
3B:
Oth
er
3C 4A 4B 4C 5A 5B 6A 6B 7 8
Small Office 193 179 181 177 156 176 146 187 176 165 202 183 228 206 246 341
Medium Office 153 145 143 139 120 137 112 138 133 118 141 133 153 139 157 204
Large Office 114 112 104 111 95 109 92 116 105 99 115 107 123 114 125 157
All 287 313 300 325 242 303 240 340 360 279 387 319 409 372 446 456

Table 2-2 Commercial Buildings Energy Consumption Survey (CBECS 2012)
Building Type EUI for the sum of major fuels (kWh/m
2
)
Overall Very cold/cold Mixed-humid Mixed-dry/hot-dry Hot-humid Marine
All buildings 252 272 257 211 233 253
Office 245 269 245 185 227 253
2.3 Simulation Tools of Electrochromic Glass
Apart from kinetic façades, dynamic glazing as an unconventional shading technique has been
studied for years. Dynamic glazing could automatically change its transparency to shield some
sunlight. Different types of dynamic glazing can be triggered by different factors. The
electrochromic (EC) glazing is controlled by electrical currents, the thermochromic glazing is
controlled by temperature, and the photochromic glazing is triggered by UV radiation (Walliams
2019). EC glass is more widely used in buildings compared with the other two. Scientists and
manufacturers have been studying it for years, including its benefits for daylighting, thermal
comfort, and energy conservation. Simulation workflows are important to analyze the
performance of dynamic shading and glazing technologies and how to improve them.
Software tools that could be used to simulate EC glazing have been studied. Bernard Paule has
tested a new possibility of simulation offered by a relatively new software DIAL+. This new tool

18
could quantify EC glazing’s performance including energy use, daylighting, and other comforts
regarding occupants (Paule 2017). EnergyPlus is a simulation engine to simulate energy
consumption including heating loads, cooling loads, electrical lighting, etc. Radiance is a
simulation engine specialized for lighting and thermal simulations developed by Greg Ward at
the Lawrence Berkeley National Laboratory (LBNL) in the 1980s (Aspen 2018). Radiance used
raytracing, which allows it to remain functional and widely used for over 40 years (Aspen 2018).
Software plugins of Rhino Grasshopper are widely used for sustainable designs in architectural
firms. Plugins like Ladybug Tools use EnergyPlus and Radiance to run simulations of daylight,
energy, and thermal comfort. Dynamic shading and glazing strategies are challenging to be
simulated because of their changing properties. The annual changes could be shown in some of
the results. Others could only be presented in the hourly results.
Yihong Cheng has developed workflows for EC glass that changes tinted levels. Cheng has
modeled two identical rooms except for glass types and used multiple software like COMFEN,
DesignBuilder, CBE Thermal Comfort Model, and Grasshopper Ladybug Tools to develop the
workflows and make comparisons. According to simulation results. CBE Comfort Model and
Grasshopper are more reliable tools than COMFEN and DesignBuilder to do EC simulations
(Cheng 2017). Ladybug Tools contain all sorts of simulations and could present data visually as
a diagram or a plot. It has been updated over the years, and the new versions contain more
functions. Therefore, the Ladybug Tools were used for the simulation process to test the overall
performance of dynamic shading and glazing strategies.

19
2.4 Parameters of Electrochromic Glass
The EC glazing has already been tested to be practical in reducing cooling loads. M. Pittaluga
has researched EC glazing as a material of a dynamic envelope that adapted to seasonal and even
daily climate changes. He also analyzed the practical use of EC glazing in a building and the
influences on energy conservation. The results showed that EC glazing has more advantages
compared with other chromogenic materials. Photochromic glazing detected solar radiation in a
visible field ranging from 380 to 780 nm, while EC glazing could also detect infrared solar
radiation ranging from 780 to 2500 nm. EC glazing was activated when the luminance reached
greater than 540 to 700 lux. The VLT of EC glazing varied from 50% (clear state) to 10% (full
state) in Pittaluga’s research, so EC glazing had a better performance in shielding solar heat and
reducing cooling loads in hot weather (Pittaluga 2015). The Sparkasse headquarters building in
Dresden constructed in 1999 had 150 m
2
of EC glazing area. It has saved 18% to 24% energy
compared with thermal insulation glazing (Pittaluga 2015). However, the EC glazing of this
research had different VLT values. Since the experiment of this research tested the city of Los
Angeles, where the climate is hot and dry, EC glazing was expected to shield sufficient sunlight
and save more energy.
Lawrence Berkeley National Laboratory (LBNL) has conducted studies of performance and
occupants’ feedback on electrochromic glass in practice. The studies included two federal
buildings retrofitted with EC glazing. The first one was the 911 Federal Building in Portland, OR
(FIG. 2-3). The second one was the Moss Federal Building in Sacramento, CA. The original 911
Federal Building used to have tinted, double-pane, low-emissivity windows, and interior blinds.
The retrofit of EC windows was installed on the third to the seventh floors of the south façade.
The building’s WWR was 46%. EC glazing for this study had four tinted levels, and the VLT

20
was 36%, 25%, 13%, and 2% respectively (FIG. 2-4). The tinted levels were controlled by
incident vertical solar radiation and could be manually overridden by occupants. Both results of
building performance and occupants’ satisfactions were positive. 36% of the electric lighting
energy was saved due to the compensation of daylighting. 2% of the cooling load of the HVAC
system was saved for the weekday mode. 57% of the cooling load was saved for the weekend
mode due to the fully tinted level of EC. 40% of the interior blinds were removed for the areas
with EC windows. While 92% of the occupants were satisfied with the renovation of EC glazing,
occupants claimed that glare was reduced, but thermal comfort had no differences from the
conventional glazing (Lee 2016). The fully tinted state had a VLT of 2%, which seemed dark,
but objects outside the window were still visible (FIG. 2-4).

Figure 2-3 South Façade of the 911 Federal Building (Lee 2016) Figure 2-4 Four Tinted Levels of EC (Lee 2016)
The Moss Federal Building (FIG. 2-5) originally had double-pane, low-emissivity windows, and
interior blinds before renovations. The EC glazing retrofit was installed on the sixth floor of the
south façade. The four tinted levels of VLT were 60%, 18%, 6%, and 1% respectively. They
were controlled by sensors on the exterior facades and determined by a central unit that was

21
based on multiple systems including solar penetration, glare, and work schedules. Each window
panel had subzones that allowed several tinted levels (FIG. 2-6). As a result, the daily HVAC
loads were decreased by 29% to 65% in different zones of the building (Fernandes 2021). The
peak HVAC loads were reduced by 25% to 58%. The energy use of electric lighting had
increased by 62%, which probably had nothing to do with EC glazing, but with other issues. No
significant glare reductions were reported by occupants. A slight reduction in interior blinds use
was reported. An improvement in thermal comfort during the hot weather was reflected. 63% of
the occupants preferred the renovation of EC glazing to conventional glazing (Fernandes 2021).
Surprisingly, while the simulation results indicated that EC glazing reduced energy consumption,
glare, and solar heat, the occupants’ feedback showed that EC glazing might not provide an
idealized performance of glare or thermal comfort. The result might be relevant to occupants’
conditions or individual issues of the buildings, but it also indicated that more potential
improvements to the adaptive shading system could be made.

Figure 2-5 South Façade of the Moss Federal Building (Fernandes 2021) Figure 2-6 Subzones of Windows (Fernandes 2021)
EC glazing could have various controllers. However, most buildings in practice tend to use
instant solar irradiance on the exterior window surface to control the tinted levels. Therefore, this

22
research also used vertical solar irradiance on the window. The sensor was placed where incident
solar irradiance was not affected by the exterior shades. When the EC glazing and exterior
shades worked together, the solar irradiance on the window surface was not be affected. When
the exterior shades were deployed, the solar irradiance on the window surface would not become
lower, and the EC glazing remained fully tinted state.
However, since two case studies had disparate VLT of EC glazing, the experiment used VLT and
SHGC values of EC glazing manufactured by SageGlass. Four states had a fixed U-value of 1.59
W/m
2
-K, but variable SHGC and VLT values. The clear state had 60% VLT and 0.47 SHGC.
The light state had 21% VLT and 0.17 SHGC. The heavy state had 6% VLT and 0.11 SHGC.
The full state had 2% VLT and 0.09 SHGC (Sbar 2012). There was also no information about
what solar irradiance would control four states of EC glazing. Therefore, the first experiment
(shoebox model) had the annual solar irradiance range divided by five. When the solar irradiance
value detected by the sensor was in the range of 0% - 10%, the EC glazing would be in its clear
state. In the range of 10% - 20%, it became the light state. When the value fell into the range of
20% - 30%, the glass turned to the heavy state. When the solar irradiance value was between
30% - 40%, the EC glass became fully tinted. When the value was in the range of 40% - 100%,
the exterior kinetic blinds were deployed. For the second experiment (the Entrada project), there
were different solar irradiance triggers given the fact that different solar irradiance was received
by glazing on different facades orientations and level heights. The new values were calculated to
make sure that different states of the dynamic systems could all be utilized. See more
information in Chapter 3.

23
2.5 Visual Comfort and Thermal Comfort Metrics
Apart from reducing cooling loads and conserving energy, EC glazing’s improvement of visual
comfort was more widely concerned. Alessandro Cannavale has researched an innovative solid-
state EC device. Energy balance has been evaluated using simulations conducted in EnergyPlus
software. Daylighting performance was assessed by Useful Daylight Illuminance (UDI) and
Discomfort Glare Index (DGI). The study stated that EC glazing could theoretically reduce
energy consumption by 52% in cooling-dominated regions, and the maximized energy reduction
could reach 40 kWh/m
2
per year in the hottest areas. The daylighting performance was reached
because UDI was increased by 82.7%, meaning more percentage of areas received illuminance
ranging from 100 lux to 2000 lux (Cannavale 2018). J. Mardaljevic and A. Nabil have discussed
the application of two façade technologies. One was to use EC glass to replace traditional glass
in commercial buildings. The second one was to install PV panels on vertical façades. The
electricity generated by PV panels would be used to compensate for the energy consumption of
electric lighting. The test model was an office with a perimeter of six meters. The research has
tested performance from four orientations, hourly weather data for a whole year, and fourteen
locations around the world. After comparing the two strategies, Mardaljevic and Nabil found
applying EC glazing alone saved more energy than installing PV panels on vertical façades.
Also, EC glazing alone could provide enough daylighting and no shading is needed (Mardaljevic
2008). However, there are different metrics of glare and some of them are not considered
appropriate for analysis. Glare analysis is an important part of daylight performance and visual
comfort. One major function of dynamic glazing and shades is to reduce the largest amount of
discomfort glare while enabling enough amount of daylight to keep the interior visual
environment bright and comfortable. To evaluate that, there are several glare metrics to choose

24
from and compare.
Professor Jae Yong Suk, Professor Marc Schiler, and Professor Karen Kensek have researched
the accuracy and consistency of the five most commonly used discomfort glare metrics. They
were Daylight Glare Probability (DGP), Daylight Glare Index (DGI), Unified Glare Rating
(UGR), Visual Comfort Probability (VCP), and CIE Glare Index (CGI) (Suk 2016). They have
analyzed more than 450 glare scenes and found DGP to be the most accurate metric among the
five metrics. However, the highest accuracy level was only over 50%, VCP and CGI were not
accurate enough to be used for glare analyses, and DGI and UGR did not perform well when
there were scenes of direct sunlight (Suk 2016). The formula of DGP included vertical eye
illumination, luminance of source, solid angle of source, and position index. DGI considered the
luminance of source, solid angle of source, background luminance, and position index. UGR and
part of the formula of CGI contain luminance of source, solid angle of source, background
luminance of source, position index, direct vertical illuminance, and indirect vertical illuminance
(Wienold 2005). VCP was different from other metrics. It measured more visual comfort and
considered less discomfort glare issues (Suk 2016). Doctor Jan Wienold has compared and
evaluated the performance of 22 glare metrics. He has also found that DGP could make the most
robust prediction of glare performance among all the metrics (Wienold 2019). This research
recorded DGP values of occupied hours on the fall equinox. The momentary results of DGP were
compared and evaluated as part of the glare analyses of different glazing and shading conditions.
The research considered possible variations in results of different climates. However, glare
analyses were not enough to evaluate occupants’ visual comfort. Spatial Daylight Autonomy
(sDA) and Annual Sunlight Exposure (ASE) were equally important metrics to assess indoor
visual comfort. Annual daylight shows the cumulative daylight results of a whole year. Spatial

25
Daylight Autonomy (sDA) is a metric that measures the percentage of a certain area that receives
sufficient daylighting which is illuminance above 300 lux for 50% of the time (sDA300lux50%)
from 8 am to 6 pm (LM 2013). This metric could be used to analyze whether a certain kind of
shading condition could provide enough daylight. Annual Sunlight Exposure (ASE) is another
annual daylight metric, which measures the percentage of direct sunlight on a horizontal surface
that provides visual discomfort (LM 2013). However, the two cumulative daylight metrics were
hard to analyze dynamic glazing and shading conditions that change states. For dynamic
conditions, average momentary illuminance values were used to evaluate whether they could
provide sufficient daylight. The percentage of areas with illuminance higher than 300 lux for
more than 50% of the occupied hours on the fall equinox was also calculated to identify daylight
performance in terms of sDA300lux50% of one day.
Jones from a built environment firm Arup has defined a new target using DGP, which was called
spatial glare autonomy (sGA). It indicated a fraction of views that had no glare, which meant that
DGP larger than 40% should not last longer than 5% of the occupied hours (sGA40%5%). A
European standard EN 17037:2018 had a target of not exceeding DGP of 45% for 5% of the
occupied time as minimum glare protection. Jones’ new target was based on EN 17037:2018 but
stricter to achieve (Jones 2019). The performance goal set for test conditions in this research was
to achieve 0% of sGA40%5%.
Thermal comfort was another essential part of interior environmental quality for the experiment.
EC glazing might shield enough sunlight visually but not thermally. It might reduce partial
cooling loads, but exterior shading could shield more unwanted heat. The combination of EC
glazing and kinetic shading might bring more benefits to both visual and thermal comfort.
Adaptive thermal comfort is a metric to evaluate the thermal conditions when the building is

26
naturally ventilated. This approach is developed based on the surveys of the thermal response by
occupants. People vote for the comfort level based on a descriptive scale, like ASHRAE.
Subjects respond whether the indoor environment is too hot, too cold, or comfortable. The
response reflects not only the indoor temperature, but also factors like humidity, air movement,
clothing, and activity. Therefore, the data of those indices are collected and used to analyze the
adaptive thermal comfort of a room instead of testing subjects for responses. Several standards
including ASHRAE 55 are based on this approach (Nicol 2002). For the experiment comparing
multiple shading and glazing strategies, the percentage of hot time of adaptive thermal comfort
metric reflected the strategies’ ability to decrease overheated conditions. However, the
commercial building in the experiment was air-conditioned instead of naturally ventilated, so
adaptive thermal comfort would not be the main metric to evaluate the thermal conditions.
For air-conditioned buildings, the metrics used to assess thermal comfort are Predicted Mean
Vote (PMV) and Predicted Percentage of Dissatisfied (PPD). They were developed by Fanger to
predict the thermal comfort or thermal discomfort of occupants. PMV is a metric rated from -3 to
+3 indicating the sensation from cold to hot. 0 means neutral, and the range from -0.5 to +0.5
stands for comfortable situations. PPD is the percentage that indicated the degree of thermal
discomfort (Koh 2018). Since most of the test conditions were active shading or glazing
strategies, and the building was air-conditioned instead of naturally ventilated in this research,
PMV and PPD were used to indicate the thermal situations of the test models. Adaptive Thermal
Comfort was used to evaluate the thermal conditions when the office building had a power
outage and only had passive shading strategies like static overhangs.

27
2.6 Full Scale Office Testbed Facilities
LBNL FlexLab has researched kinetic shading systems with built facilities to test energy and
lighting savings. To research the performance, simple test facilities with windows on one façade
have been built. Sensors and manikins were placed in the test facilities (Carbonnier 2021). To
test the real effect of shading and glazing systems on solar heat and sunlight, the physical test
facilities were very useful. For the experiment of this thesis, test facilities with different shading
and glazing options were not built because of the limited budget. However, a simple shoebox
model with windows on one façade was tested first. Then a standard model of an existing large
office building was tested. The standard office building had window areas on all four sides, so
the conditions were different from a shoebox model. Also, the experiment used SI units instead
of IP units.
2.7 Conclusions
More than a thousand studies have been done about the building performance of kinetic façades
and dynamic glazing according to Google Scholar. Different digital tools and models were used
for simulation, different benchmarks and standards were used for calculations and baselines, and
different building performance was concerned. Different shading and glazing techniques all have
offered environmental benefits and improved multiple building performance and occupants’
comfort at the same time. However, there was space for improvement. There was no previous
study about the performance of combinatorial shading and glazing technologies. This research
showed the combinatorial strategy had superior results compared to other shading and glazing
techniques. The literature review helped the experiment to determine that kinetic miniature
blinds, kinetic horizontal blinds, and static overhang would be tested. Incident solar irradiance on

28
the window would be the external stimuli for the dynamic shading and glazing system. VLT
values for the EC glazing would be 60%, 21%, 6%, and 2%. Metrics of EUI, DGP, PMV, and
PPD would be used. Also, peak cooling loads and average momentary illuminance larger than
300 lux were used to evaluate the energy and visual performance as well. The test conditions
would be static clear glazing, clear glazing with kinetic horizontal blinds, clear glazing with
kinetic miniature blinds, clear glazing with static overhang, EC glazing, EC glazing with kinetic
horizontal blinds, EC glazing with kinetic miniature blinds, and EC glazing with static overhang.
The first experiment was done on a shoebox model, and then workflows were applied to a real
office building model in the second experiment.

29
CHAPTER THREE METHODOLOGY
According to the literature review, the performance of static clear glazing, dynamic glazing,
static shading, kinetic shading, and combinatorial systems should all be tested and compared
systematically. In order to test whether a combined system with both EC glazing and kinetic
shading could have better performance in energy, daylight, glare, and thermal comfort over test
conditions with other glazing and shading techniques, a series of simulations were conducted,
and a series of test conditions were compared (FIG. 3-1). The experiments were to simulate
daylight, glare, peak HVAC cooling loads, EUI, PMV thermal comfort, and PPD of the eight
conditions. The eight test conditions were static clear glazing, static glazing with kinetic
horizontal blinds, static glazing with kinetic miniature blinds, static glazing with static overhang,
EC glazing, EC glazing with kinetic horizontal blinds, EC glazing with kinetic miniature blinds,
and EC glazing with static overhang. A scoring system was developed based on the performance
indicators and performance goals. The test condition receiving the highest score was the most
effective technique among other test conditions.
The reason for doing simulations instead of manual calculations based on reality was that there
were too many variables that could affect the building's performance. Even for the same building
type with the same floor areas in the same climate zone, factors like building schedules,
occupancy, and mechanical systems could result in significant differences in energy, visual, and
thermal performance. Since only the influence of the façade system was tested, a software
simulation program was used to eliminate the nuance brought by other factors.

30

Figure 3-1 Methodology Overview
3.1 Analyze Existing Shading Technologies
Before conducting experiments to test the building performance under several test conditions, the
performance of existing buildings has been researched in Chapter 2. Developers and architects
tend to publicize the energy data if the kinetic shading or EC glazing reduced the environmental
impacts. For example, sustainable buildings like Bullitt Center and Santa Monica City Service
Building have publicized their EUI. Average EUI data of the regions were used as the baseline to
compare. LBNL’s EC glazing studies of the 911 Federal Building and the Moss Federal Building
have also shown reduced peak cooling loads and electric lighting. However, other buildings with
static glazing usually do not have available data of energy use or other building performance like
reduced artificial lighting or HVAC loads. The claim of large office buildings with curtain wall
façades consume large amounts of energy is speculation. To demonstrate the claim, it is
necessary to conduct energy simulations of the same buildings with different types of glazing
and shading. Although the results varied from the actual conditions because of the complexity of
reality, they were still valid enough to support or refute the claim.

31
3.2 Preliminary Shoebox Model Text [Experiment #1]
After the existing buildings have been researched, the dynamic shading and glazing system was
modeled and simulated on software programs. Before doing experiments on a standard-sized
model, it was important to test all the test conditions by using a simple shoebox model with
simple workflows. That was Experiment #1. As mentioned before, both EnergyPlus and
Radiance were valid simulation engines that supported the Ladybug Tools of Grasshopper.
Therefore, the Ladybug Tools were tested for the shoebox model first and then used for the
standard office building model.
3.2.1 Select Software Programs
Grasshopper is a software tool of modeling tool Rhino, and it is widely used as a simulation tool
among architectural firms for sustainable design. When the architecture team has done the model
design in Rhino, the sustainable design consultants use Grasshopper plugins to test whether the
design has met the requirements of sustainability. There are lots of grasshopper plugins that
could do energy and lighting simulations. Except for the Ladybug Tools, DIVA and
ClimateStudio could also be used for daylight and energy analysis. For this experiment, the
Ladybug Tools were used because they could simulate energy and thermal performance of the
EC glazing and kinetic blinds.
3.2.2 Model a Simple Room with Windows
A small room with a window on one side of the wall was used to test the initial workflow. It was
clear to see whether the performance results from the workflows would be reliable or not for the
office building test because the shoebox model has all the components of a building. It did not
take too much time to simulate because the model was simple. The simulation software is easy to

32
crash due to the complexity of the model, data, and simulation process. A shoebox model
simplified the process, so it was efficient to test the workflow before it was fully developed.
When changes had to be applied to the workflow, it was also convenient to be tested on the
shoebox model first.
The shoebox model (FIG. 3-2) represented a small corner from a floor of an office building. It
had one exterior wall with an aperture and three interior walls. Research regarding daylighting
showed the most efficient WWR for different climates was tested to be 30% to 45% (Sayadi
2021). Therefore, the model had a WWR of 30% as a baseline. The window was located on the
west façade so the change of sunlight through the window was obvious. The shoebox model had
length of 6 meters, width of 8 meters, and height of 3 meters. A sensor was placed on the
window surface to detect the solar irradiance to control dynamic glazing and shading.

Figure 3-2 Shoebox Model
3.2.3 States Change During a Day
One of the most essential parts of the proposed dynamic systems was the dynamic property, so
the VLT of the EC glazing should be editable. When there was an input option of VLT values, it
could be controlled by solar vertical irradiance on the exterior window side. The kinetic shade
was controlled by the same factor with a different setpoint. When the states of glazing and
shading were able to change automatically during the day, the basic workflow of the dynamic

33
shading and glazing system was done.
3.2.4 Simulate Visual, Energy, and Thermal Results
The simulation results of hourly illuminance and glare were expected to be different when the
VLT of the glass changed or when the blinds were deployed. The VLT of clear glass and tinted
glass should lead to significant variations in daylight results. The deployed blinds should result
in obvious differences in thermal comfort and energy reduction. If the variations in results were
not obvious, it would be hard to test the performance of the test conditions. With the conspicuous
changes in simulation results, adjustments could be made.
See below for the information on the shoebox model (Table 3-1). Chapters 2 and 3 have
explicitly explained why these values were used. A formula of VLT values changing with the
vertical solar irradiance on the window was created using the Ladybug Tools in Grasshopper.

34
Table 3-1 Information of the Shoebox Model
Shoebox Model
Dimension 6 m * 8 m * 3 m
WWR 30%
City Los Angeles
Climate mild-to-hot, mostly dry
Controller Solar irradiance on the window surface (sensor)
EC Clear State VLT 0.60
U value 1.59 W/m
2
-K
SHGC 0.47
Setpoint ≥ 0 W/m
2

EC Light State VLT 0.21
U value 1.59 W/m
2
-K
SHGC 0.17
Setpoint ≥ 10% of the highest solar irradiance of a year
EC Heavy State VLT 0.06
U value 1.59 W/m
2
-K
SHGC 0.11
Setpoint ≥ 20% of the highest solar irradiance of a year
EC Full State VLT 0.02
U value 1.59 W/m
2
-K
SHGC 0.09
Setpoint ≥ 30% of the highest solar irradiance of a year
Kinetic Blinds ≥ 40% of the highest solar irradiance of a year
3.3 List of Test Conditions
Since different conditions of glazing and shading were compared, it was important to consider
what kinds of situations were included in this research. According to Chapter 2, the performance
of static glazing, dynamic glazing, static shading, kinetic shading, and combinatorial systems
should all be tested and compared systematically. The assumption was that combining dynamic
glazing and kinetic exterior shading led to the best results of comprehensive performance of
energy, visual, and thermal comfort. The table shows information of all eight test conditions
(Table 3-2).

35
Table 3-2 Test Conditions of the Shoebox Model
Clear Double-Pane Glazing: U-value: 1.69 W/m
2
-K, Solar Heat Gain Coefficient (SHGC):
0.42, Visible Light Transmittance (VLT): 0.64
Test Condition #1 Test Condition #2 Test Condition #3 Test Condition #4
Double-Pane
Glazing (D)
Kinetic Horizontal
Blinds (D + H)
Kinetic Miniature
Blinds (D + M)
Static Overhang (D
+ O)
Length: 4.38 m
Height: 1.64 m

Slat Width: 0.1 m
Spacing: 0.164 m
Tilt Angle: 0°
Same aspect ratio as
the overhang
Slat Width: 0.03 m
Spacing: 0.02 m
Tilt Angle: 45°
Regular blinds
Different aspect ratios
Depth: 1 m
Electrochromic (EC) Glazing: U-value: 1.59 W/m
2
-K
Clear State: 0.60
VLT, 0.47 SHGC
Light State: 0.21 VLT,
0.17 SHGC
Heavy State: 0.06
VLT, 0.11 SHGC
Full State: 0.02
VLT, 0.09 SHGC
Test Condition #5 Test Condition #6 Test Condition #7 Test Condition #8
EC Glazing (E) EC + Kinetic
Horizontal Blinds (E
+H)
EC + Kinetic
Miniature Blinds (E
+M)
EC + Static
Overhang (E + O)
3.3.1 Clear Glazing Without Exterior Shading [Test Condition #1]
Test condition #1 was clear double-pane glazing in the EnergyPlus material library. It was a
code-compliant glazing type and could be used as a reference glazing type. Its SHGC was 0.35,
VLT was 0.64, and U-factor was 1.69 W/m
2
-K. This test condition had static glazing and had no
exterior shading techniques. The window was not operable. Its length was 4.38 m and its height
was 1.64 m (FIG. 3-3).
3.3.2 Kinetic Blinds [Test Conditions #2 & #3]
Test conditions #2 and #3 were clear glazing working with exterior kinetic blinds. Two types of
blinds with different aspect ratios and tilt angles were tested. Test condition #2 was horizontal
blinds with slat widths of 0.1 m, spacings of 0.164 m, and 0° of tilt angles (FIG. 3-4). Sunlight
could still go through the windows when horizontal blinds were deployed. Test condition #3 had

36
typical indoor miniature blinds outside the windows. The miniature blinds had slat widths of 0.03
m, spacings of 0.02 m, and tilt angles of 45° (FIG. 3-5). When the miniature blinds were
deployed, most of the daylight and glare were shielded. They could also block unwanted solar
heat since they were put outside the windows. The blinds would be lowered when the incident
solar irradiance values on the window fell into 40% - 100% of the annual range.
3.3.3 Static Overhang [Test Condition #4]
An overhang that extruded 1 m from the window top is the fourth test condition (FIG. 3-6) for
the shoebox model. The overhang had the same aspect ratio and tilt angle as the horizontal
blinds, which was 1/1.64, so the overhang should have the same performance as horizontal
blinds were deployed. However, horizontal blinds were kinetic, while the overhang was fixed.
Therefore, test conditions #2 and #4 had similar performance when they were both deployed, but
different annual performance. The overhang of Experiment #2 extruded 2.63 m because the
window height was 4.31 m. The aspect ratio then stayed the same.

Figure 3-3 Test Condition 1 Figure 3-4 Test Condition 2 Figure 3-5 Test Condition 3 Figure 3-6 Test Condition 4

37
3.3.4 EC Glazing
EC glazing was the fifth test condition and had properties of EC glazing manufactured by
SageGlass as a reference (FIG. 3-7). Four states of the EC glazing had a fixed U-value of 1.59
W/m
2
-K, and variable SHGC and VLT. The clear state had 60% VLT and 0.47 SHGC. The light
state had 21% VLT and 0.17 SHGC. The heavy state had 6% VLT and 0.11 SHGC. The full
state had 2% VLT and 0.09 SHGC (Sbar 2012). The switching setpoints of light, heavy, and full
states were 10%, 20%, and 30% of the highest value of solar irradiance of the year detected by
the sensor on the window surface.
3.3.5 Combined Systems of EC Glazing and Kinetic Blinds [Test Conditions #6 & #7]
Test conditions #6 and #7 had an extra step of lowering the automated exterior blinds when the
sensor detects 40% of the highest solar irradiance of the year on the window surface. When the
incident solar irradiance was below 40% of the highest annual value, they were lifted again. In
order to control variables, test condition #6 (FIG. 3-8) had the same kinetic horizontal blinds as
test condition #2, and test condition #7 (FIG. 3-9) had the same kinetic miniature blinds as test
condition #3. Exterior blinds should shield the sunlight better than fully tinted glass. The
combinatorial strategies were expected to further reduce the glare and overheated conditions
caused by unwanted solar heat gain compared to using dynamic glazing or kinetic shading alone.
3.3.6 Combined System of EC Glazing and Static Overhang
The last test condition placed a static horizontal overhang over an EC window (FIG. 3-10). This
was based on the assumption that the overhang blocked the unwanted solar heat while dynamic
glazing could provide visual comfort. Although kinetic blinds were usually considered more
effective to shield solar heat and glare, overhang with the same aspect ratio could provide the

38
same effect while enabling more daylight. Again, test condition #8 had the same overhang as test
condition #4 to control variables.

Figure 3-7 Test Condition 5 Figure 3-8 Test Condition 6 Figure 3-9 Test Condition 7 Figure 3-10 Test Condition 8
3.4 Determining Performance Indicators
All test conditions were evaluated based on various performance indicators, including energy,
visual, and thermal aspects. If a building was only aimed at low energy consumption, it could
have inferior performance to other indoor environmental qualities. For example, if there was less
window area, there would be less solar heat gain, thus less energy would be consumed for
cooling loads. However, with less window area, there would be less amount of daylight, and
more energy consumed for electrical lighting. Hence, only when all the indicators embody
acceptable performance, the test conditions could be considered effective.
3.4.1 Visual Comfort
The first performance category to evaluate was visual comfort. The expectation was that the
dynamic glazing was the most responsible for that. It was the same role as the interior shading
system of the Bullitt Center. Tinted glazing might reduce thermal heat gain to some extent, but

39
the exterior shading system was more responsible for that. However, the exterior shading system
should also reduce the discomfort glare while not shielding useful daylight.
3.4.1.1 Daylight
The property of the dynamic shading and glazing system was that they change states. The
quantitative changes of dynamic systems could be shown by “Small Multiples”, which
documented the daylight simulation results hour by hour for one day in this case. “Small
Multiple” proposed by Edward Tufte was a strategy showing a series of graphs with the same
scale (Gunter 1997). This point-in-time metric simulated the illuminance (reported in lux) of the
working surface level of the room hour by hour. When kinetic façades and EC glazing were
assigned to a schedule, simulation results of different hours of the day clearly reflected shading
and glazing systems’ contribution to the building performance.
Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE) values were not
available for dynamic glazing and shading devices. However, a rough percentage of areas with
illuminance larger than 300 lux could be calculated with the hourly results of indoor illuminance
of the working surface during the day. Although the results were not accurate compared to real
conditions, they could be useful references. The autumn equinox was used to represent the whole
year’s daylight performance because the sun positions were in the middle compared with the
extreme positions of summer and winter solstices. The percentage of areas with illuminance
larger than 300 lux of each hour from 7 am to 6 pm was calculated and then the average value
was taken.
Additionally, a scoring system was used to determine which test condition could earn a higher
score. In this case, a higher value of the average percentage of areas with illuminance larger than

40
300 lux on the fall equinox had a higher score because the goal is to maximize the daylight and
achieve the goal of transparency.
3.4.1.2 Glare
Glare was an important factor in visual performance. While different kinds of glazing and
shading systems were providing enough daylight, they should also be able to prevent discomfort
glare. Different glare metrics have been discussed and explained in Chapter 2. Daylight Gare
Probability (DGP) was the main metric to evaluate the glare performance of different test
conditions. The strategy of “Small Multiple” was again used to show the variances of sky
luminance at different hours during the day. The test condition with a higher value of average
DGP earned a higher glare score.
3.4.2 Thermal Comfort
The thermal comfort metric included Predicted Mean Vote (PMV) thermal comfort developed by
ASHRAE 55, which predicted percentages of hot, cold, and comfortable time according to
occupants’ sensations in air-conditioned buildings, and Predicted Percentage Dissatisfied (PPD),
which predicted the level of thermal discomfort in air-conditioned rooms.
3.4.2.1 PMV Thermal Comfort
PMV thermal comfort was used to measure the percentage of thermal conditions that were
considered comfortable, too hot, or too cold for the occupants. It was calculated based on
occupants’ thermal sensations from -3 to +3, which meant too cold to too hot (Koh 2018). To
measure that, the room was assumed to be air-conditioned instead of naturally ventilated, which
aligned with situations that office buildings usually have mechanical cooling and heating

41
systems. The less percentage of the overheated zone over a year meant the shading or glazing
system was more effective for providing a thermally comfortable indoor environment.
Overheated zone meaning the sensation points were larger than +0.5. Therefore, a lower
percentage of overheated conditions indicated a higher PMV thermal score.
3.4.2.2 PPD
PPD showed degrees of thermal discomfort over a year. The hourly plot directly showed the
levels of uncomfortable thermal conditions that occupants could feel. Hence, it was
straightforward to see whether the test conditions were effective enough to provide a comfortable
thermal environment in the interior space. It could also show whether the shading system kept
the indoor thermal conditions within the range recommended by ASHRAE Standard 55-2017.
PPD values ranged from 5 to 100%. Values less than 10% meant comfortable. Values lower than
20% were considered efficient. There was no distinction between cold and hot conditions (Koh
2018). The smaller the PPD value of the hot conditions, the higher the temperature score.
3.4.3 Energy Use
The metrics for energy use contain a momentary one and a cumulative one. The peak cooling
load measured the highest cooling load on the summer design day which was the day with the
highest amount of cooling load the HVAC system uses. Energy Use Intensity (EUI) was another
metric that focused on the annual energy consumption of a building including elements of
cooling, heating, artificial lighting, and interior equipment.
3.4.3.1 Peak Cooling Loads
The peak cooling load [reported in Watts (W)] was an important metric to determine the ability

42
of different test conditions to reduce energy for the hottest day of the year. The test conditions
with EC glazing or kinetic blinds should be effective for the reduction of peak cooling loads
because they shielded sunlight to decrease solar heat gain. The test condition with the lowest
peak cooling loads won the highest cooling score.
3.4.3.2 EUI
With known annual loads, EUI [reported in kilowatt-hours per square meter (kWh/m
2
)] could
also be calculated and to be compared with benchmarking EUI. However, not every sector of the
EUI was accurately calculated because simulations cannot perfectly reflect real life. For
example, how the shading and glazing affected the use of equipment and occupancy was not be
considered. The main part was the energy consumption of annual heating, cooling energy usage,
and electric lighting. The simulation software provides a fixed interior lighting amount for all
test conditions. The daylight harvesting strategy was used because natural light providing
illuminance larger than 300 lux could compensate for the artificial lighting. Therefore, the
amount of artificial lighting multiplying with the average percentage of areas with illuminance
larger than 300 lux was reduced from the EUI value. Not increasing too much heating EUI was
equally important as reducing annual cooling EUI, especially for heating-dominated regions,
because the ultimate goal was to have an overall lower energy consumption. Even in Los
Angeles, a cooling-dominated city, exterior shades could lead to a reduction in cooling EUI but
an increase in heating EUI. The effective test conditions should be able to create a balance and
conserve energy after all. A lower EUI value had a higher EUI score.

43
3.4.4 Performance Scoring System
With the scores of daylight, glare, peak cooling, EUI, PMV, and PPD, the most effective shading
technique based on performance could be assessed. Each of the categories could weigh
differently. If glare was the most important issue for the designer, it could weigh more than other
factors. If energy was the most significant indicator for the designer, the EUI score could weigh
ten times more than other ones.
For this experiment, visual comfort, thermal comfort, and energy use had the same weight.
Visual comfort contained an average percentage of areas receiving illuminance larger than 300
lux and the average values of DGP. Both of the performance indicators were scored from one to
eight. Thermal comfort contained the percentage of PMV overheated time and PPD of
overheated conditions, and they were also scored from one to eight. Energy use contained
peaking cooling loads and EUI, and both of them score from one to eight. Each of the
performance indicators had a rank indicating which test condition had the best individual
performance.
A scoring sample below (Table 3-3) shows how the final scoring system worked. The scores
were just sample scores that did not mean anything. All categories had the same weight and
scored from one point to eight points. The overall points were added together, and Test
Condition #8 had the highest score, which meant it was considered the most effective shading or
glazing strategy among all.

44
Table 3-3 A Scoring Sample of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Daylight Score 1 2 3 4 5 6 7 8
Glare Score 1 2 3 4 5 6 7 8
Peak Cooling Score 1 2 3 4 5 6 7 8
EUI Score 1 2 3 4 5 6 7 8
PMV Thermal Score 1 2 3 4 5 6 7 8
PPD Score 1 2 3 4 5 6 7 8
Overall Performance Score 6 12 18 24 30 36 42
48☆
Overall Performance Rank 8 7 6 5 4 3 2 1
More importantly, the highest-ranked test condition could be different from the real model test.
The shoebox model only had one window on the west façade, so the daylight score of test
conditions with shading was much lower than other conditions. Real glass office buildings with
all-glass façades provided much more daylight and glare. For the same reason, the thermal score
of the test conditions with dynamic glazing and kinetic shading would be lower than expected.
While reducing hot conditions, they would create more cold conditions and result in less
comfortable conditions because of the limited solar heat from the west window. The glass office
building would have a lot hotter conditions for the static glazing condition.
3.5 Model a Real Office Building [Experiment #2]
Modeling an existing office building was Experiment #2 of the research. Currently, large office
buildings are popular among developers. Large proportions are glass buildings or buildings with
large proportions of window area because they provide sufficient daylight and incorporate
aesthetics (Love 2019). However, glass leads to significant energy consumption because of the
solar heat gain. Also, a large office tends to have numerous occupants and most of them are
working day and night, so the occupant’s comfort is important. Improvements of technologies
made to façades of large glass office buildings have large contributions.

45
3.5.1 The Entrada Project
The standardized model should be able to represent the most typical glass office buildings
constructed by the industry. Large office buildings usually have a high WWR because of
aesthetic reasons. Glass façades do not only provide sufficient daylight for occupants but also
make the buildings have great exterior appearances.
Entrada was a case study of a glass building with no exterior shading devices. There was no
energy data available, but it would consume significant energy by speculation because of its
form, materials, and its location. Large glass areas would lead to solar heat gain and create too
much internal solar heat even in winter. It might have a very high cooling load, so the peak
cooling loads and EUI based on minimum code requirements were simulated by using software
simulation programs. All test conditions were applied to the building.
The Entrada project was modeled based on the scale found on Google map (FIG. 3-11) and its
internal plan (FIG. 3-12). It was a complicated building model, and this would take longer to
simulate and analyze the performance. Therefore, only the southwest corner of the third level of
the building was used for simulations (FIG. 3-13). The corner had windows on the west and
south facades, and the east and north sides were partitioned walls. The red dots on the facades
were sensors of solar irradiance. Arrows meant orientations of sensors. The mullions on the
window and columns were considered part of the simulations, but the furniture was modeled but
not included in the simulation process due to the complexity. Also, furniture was moveable, so it
would not have too much impact on the building's performance.

46

Figure 3-11 The Full Site of Entrada with Scale

Figure 3-12 The Third Floor Plan and Model of Entrada (LPC West 2021)

Figure 3-13 Southwest and Northeast Isometrics of the Third Floor Corner of Entrada
The dimension of the model was 22.27 meters * 13.76 meters * 4.31 meters. It was on the third
floor. The Entrada building was built on another five-level parking structure, so the floor of the
Sensors

47
model was 30.10 meters above the ground. The two red dots on the facades were sensors for
incident solar irradiance. The table below shows more information of the test conditions of the
Entrada project, which was a little different from the shoebox model (Table 3-4).
Table 3-4 Information of Test Conditions of the Entrada Project
Clear Double-Pane Glazing: U-value: 1.69 W/m
2
-K, Solar Heat Gain Coefficient (SHGC):
0.42, Visible Light Transmittance (VLT): 0.64
Test Condition #1 Test Condition #2 Test Condition #3 Test Condition #4
Double-Pane
Glazing (D)
Kinetic Horizontal
Blinds (D + H)
Kinetic Miniature
Blinds (D + M)
Static Overhang (D
+ O)
West Length: 13.76
m
South Length:
22.27 m
Height: 4.31 m

Slat Width: 0.1 m
Spacing: 0.164 m
Tilt Angle: 0°
Same aspect ratio as
the overhang
Slat Width: 0.03 m
Spacing: 0.02 m
Tilt Angle: 45°
Regular blinds
Different aspect ratios
Depth: 2.63 m
Electrochromic (EC) Glazing: U-value: 1.59 W/m
2
-K
Clear State: 0.60
VLT, 0.47 SHGC
Light State: 0.21 VLT,
0.17 SHGC
Heavy State: 0.06
VLT, 0.11 SHGC
Full State: 0.02
VLT, 0.09 SHGC
Test Condition #5 Test Condition #6 Test Condition #7 Test Condition #8
EC Glazing (E) EC + Kinetic
Horizontal Blinds (E
+H)
EC + Kinetic
Miniature Blinds (E
+M)
EC + Static
Overhang (E + O)
3.6 Controllers of Dynamic Systems
Since the case studies of EC glazing in practice have not stated the setpoints of incident solar
irradiance that switched states of EC glazing, the range of annual solar irradiance on the vertical
exterior window surface was divided into five ranges for four states of EC glass and kinetic
blinds. Light, heavy, and full states were switched when the solar irradiance reached 10%, 20%,
and 30% of the highest annual value. The kinetic blinds were lowered when the solar irradiance
reaches 40% of the highest annual value. The sensors were put on the exterior window surfaces
pointing horizontally outwards. If the highest incident solar irradiance of the year was 800 W/m
2
,

48
the EC glazing would switch from the clear state to the light state when the solar irradiance on
the window was 80 W/m
2
. The kinetic blinds would be deployed when the solar irradiance was
320 W/m
2
.
The Entrada Project had a different algorithm of solar irradiance of dynamic shading and glazing
states due to different solar conditions on the south and west facades. More information on solar
irradiance and the Entrada Project were shown in the table below (Table 3-5). The values
guaranteed that different states of dynamic systems would all be utilized.
Table 3-5 Information of the Entrada Project
The Entrada Project
Dimension 22.27 m * 13.76 m * 4.31 m (30.10 m above the ground)
WWR 95% on both west and south facades
City Culver City, Los Angeles County, CA
Climate mild-to-hot, mostly dry
Controller Solar irradiance on the window surface (sensor)
EC Clear
State
VLT 0.60
U value 1.59 W/m
2
-K
SHGC 0.47
Setpoint ≥ 0 W/m
2

EC Light
State
VLT 0.21
U value 1.59 W/m
2
-K
SHGC 0.17
Setpoint ≥ 4% of the highest solar irradiance of a year on the west façade
≥ 8% of the highest solar irradiance of a year on the south façade
EC Heavy
State
VLT 0.06
U value 1.59 W/m
2
-K
SHGC 0.11
Setpoint ≥ 7% of the highest solar irradiance of a year on the west façade
≥ 15% of the highest solar irradiance of a year on the south façade
EC Full
State
VLT 0.02
U value 1.59 W/m
2
-K
SHGC 0.09
Setpoint ≥ 11% of the highest solar irradiance of a year on the west façade
≥ 23% of the highest solar irradiance of a year on the south façade
Kinetic
Blinds
≥ 15% of the highest solar irradiance of a year on the west façade
≥ 30% of the highest solar irradiance of a year on the south façade

49
3.7 Performance Goals
The final step was to evaluate the results. The simulation results of Experiments #1 and #2 were
compared with the performance goals. Combinatorial systems like test conditions #6, #7, and #8
were expected to meet most performance targets and standards.
3.7.1 Energy
The first goal was to accomplish Architecture 2030, reducing energy use by 80% of the regional
average. The benchmarks of energy use varied a lot among building types and climate zones.
There were also different benchmarking data of the energy use of certain building types and
climate zone that could be used as references. There were NREL Evaluation of ASHRAE 90.1-
2009 and CBECS 2012 mentioned in Chapter 2. They provided regional average EUI data,
which indicated whether the test conditions could help the buildings conserve enough amounts of
energy. Data from CBECS was chosen because it was real EUI data from surveys. The further
step was to evaluate whether buildings with different test conditions could have energy use 80%
lower than the same building type of the regional average. The regional average data here was
the energy data of the same building type in the same climate zone. The office building type in
the hot-dry climate had a EUI of 185 kWh/m
2
. 80% lower meant the target was 37 kWh/m
2
.
3.7.2 Visual Comfort
WELL Building Standard v1 recommended buildings to have at least 55% of sDA, and less than
10% of ASE. Although sDA was not accessible for the dynamic shading or glazing system, 50%
occupied time with the percentage of areas receiving illuminance larger than 300 lux on one day
could still be analyzed. This indicator could be called “sDA of the fall equinox”, which did not
represent the daylighting performance of the whole year but was still meaningful. Daylighting

50
was an important factor that indicated indoor environmental comfort. Energy, visual, and thermal
comfort had trade-offs. When one target was to keep energy low, test conditions need to
maintain daylight at a certain level, so that there were no extra loads for electrical lighting. Also,
when 300 lux was met for most of the working area, the test conditions need to make sure the
working space did not have a disturbing bright light that hurts occupants’ eyes. Both glares on
the vertical windows and sun exposure on the horizontal surfaces should be kept at an
imperceptive level. The goal was to create a visual comfort working condition for occupants. The
performance goal of glare was 0% spatial glare autonomy (sGA40%5%) introduced in Chapter 2.
DGP larger than 40% could not happen for more than 5% of the occupied hours (Jones 2019).
This meant the occupied hours should not have any disturbing or intolerable glare situations.
3.7.3 Thermal Comfort
Thermal comfort was an equally important factor that indicated indoor environmental quality.
While ASHRAE 55 has regulated PMV and adaptive thermal comfort, the temperature was a
more straightforward metric to reflect occupants’ comfort level. The temperature range
recommended by ASHRAE Standard 55-2017 was from 67 °F to 82 °F (19.4 °C to 27.8 °C),
which were comfortable indoor temperatures for humans (ANSI/ASHRAE 2017). When the
indoor temperature was too high or low, extra cooling or heating loads would be generated by
the HVAC system. Therefore, the test conditions need to keep the solar heat out when the indoor
temperature was too high and to let the solar heat in when the indoor temperature was too low.
3.7.4 View Quality
View quality was not in the scoring system because it was hard to be quantified. It was important
to windows but hard to evaluate based on a certain standard. Hence, view quality was considered

51
a performance target to evaluate the test conditions. Won Hee Ko has written about the
assessment of view quality of windows. View content, view access, and view clarity were three
variables to define view quality (Ko 2021). For this research, view content was not evaluated
since the content outside the window was not part of the experiment. View access and view
clarity were used for evaluation. For instance, when blinds were deployed all the time, occupants
could not access the view through the window, but kinetic blinds achieved view access because
there were not deployed for several hours of a day. When EC glazing was tinted, the view was
less clear compared with the view through clear static glazing, but EC glazing still achieved view
clarity there were hours of the clear state.
3.8 Summary
Conducting a clear methodology was essential for the experiments. The first step was to do case
studies of buildings with different existing shading and glazing techniques in practice. The
second step was to make a simple shoebox model to develop the workflows of each of the test
conditions. When all the test conditions could be simulated without errors, it was time to model a
standard-sized office building and apply the workflows to it. Then the performance results of test
conditions were compared. The scoring system was developed. The test condition with the
highest scores indicated it was the most effective shading or glazing system regarding all
performance of energy usage, daylight and glare, and thermal comfort. The final step was to
evaluate whether the test conditions met the performance goals. The performance results and the
scoring system could be meaningful to architects and engineers to make decisions in the design
process.

52
CHAPTER FOUR RESULTS
A series of simulations of eight test conditions have been done. First, the experiment of the
shoebox model has been done, which was referred to as Experiment #1. The data has been
collected. Then the experiment of the Entrada project has been done, which was referred to as
Experiment #2. The results were different from the ones of the shoebox model. However, the
latter one was closer to the real-life situation.
This chapter records the simulation results of daylight, glare, peak cooling loads, Energy Use
Intensity (EUI), Predicted Mean Vote (PMV) thermal comfort, and Predicted Percentage of
Dissatisfied (PPD). Chapter 5 will analyze the data and discuss the ranking rubrics. The superior
strategy with shading and glazing was the test condition with the highest score in Experiments #1
and #2.
Abbreviations for eight test conditions were used in the chapters displaying simulation results
and analyses (Table 4-1).
Table 4-1 Abbreviations of Test Conditions
Test
Conditions
#1 #2 #3 #4
Full Names Double Generic
Pane Glazing
Double Generic
Pane Glazing +
Kinetic Horizontal
Blinds
Double Generic
Pane Glazing +
Kinetic 45°
Miniature Blinds
Double Generic
Pane Glazing +
Static Overhang
Abbreviation D D+H D+M D+O
Test
Conditions
#5 #6 #7 #8
Full Names Electrochromic
Glazing
Electrochromic
Glazing + Kinetic
Horizontal Blinds
Electrochromic
Glazing + Kinetic
45° Miniature
Blinds
Electrochromic
Glazing + Static
Overhang
Abbreviation E E+H E+M E+O

53
4.1 Experiment #1 - Shoebox Model
The image (FIG. 4-1) below was the shoebox model. The first half of the chapter records
simulation data of the shoebox model. The data was presented in tables and diagrams.

Figure 4-1 Shoebox Model
4.1.1 Daylighting
The table (Table 4-2) records the states of glazing and shading on the fall equinox when the test
conditions included dynamic systems. The changing states applied to the daylighting and glare
sections.
Table 4-2 States of Dynamic Systems on 9/21
Time \ Facade 7 am 8 am 9 am 10 am 11 am 12 pm
West Clear Clear Clear Clear Clear Light
Time \ Facade 1 pm 2 pm 3 pm 4 pm 5 pm 6 pm
West Heavy Full Full &
Blinds
Full &
Blinds
Heavy Clear
Daylight availability was the first performance goal being tested. Daylight data on the working
surface (1 meter above the floor) has been recorded hourly from 7 am to 6 pm on September 21
st

(fall equinox) as “Small Multiples” of momentary illuminance graphs. “Small multiples” were a
Solar
Irradiance

54
series of graphs with the same scale. This graphical technique was proposed by Edward Tufte,
and the term should be described as “multiple small graphs” (Gunter 1997). It could clearly
display the change of daylight and glare performance hour by hour. In the meantime, the
percentage of area with illuminance larger than 300 lux for each hour was also recorded. This
meant sufficient daylighting was received. The table (Table 4-3) below shows the average
percentage area larger than 300 lux for each of the test conditions, and sDA on the fall equinox,
which meant 50% time of occupied hours received sufficient daylighting. See Chapter 5 for
further data analyses. The scale of the momentary illuminance was 0 to 300 lux (FIG. 4-2).
Table 4-3 Daylighting Data of All Test Conditions
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% ≥ 300 lux on
9/21
24.31% 23.35% 18.23% 16.88% 13.37% 11.72% 9.55% 6.16%
sDA300,50%
on 9/21
33.77% 32.12% 27.17% 25.52% 17.45% 15.71% 14.06% 9.64%

Figure 4-2 Scale of Illuminance
4.1.1.1 Test Condition #1 (D)
The first test condition had static code-compliant double-pane glazing and had no exterior
shading system. Therefore, test condition #1 had more daylight availability than other test

55
conditions (FIG. 4-3). It could also lead to a more uncomfortable glare, but it would be discussed
later. Since the only window of the shoebox model was located on the west façade, a significant
amount of solar heat gain came from the west window in the afternoon. There was more
percentage of working surface area with illuminance larger than 300 lux in the afternoon. At 5
pm, there was 57% of the area with larger than illuminance of 300 lux. For the entire day, the
average percentage of areas with illuminance larger than 300 lux was 24.31%. The sDA on the
fall equinox was 33.77%.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-3 Momentary Illuminance of Test Condition #1

56
4.1.1.2 Test Condition #2 (D+H)
The second test condition was double-pane glazing with exterior kinetic horizontal blinds
system. The blinds were deployed when incident solar irradiance on the window was high. For
the windows on the west façade, it happened in the afternoon, around 3 and 4 pm. Therefore, the
percentage of the working surface area larger than 300 lux was reduced in the afternoon (FIG. 4-
4). However, when the sun angle was low at 5 pm, the incident solar irradiance on the window
was lower than 3 and 4 pm, so the blinds were not deployed. There was still a large percentage of
the area with illuminance larger than 300 lux. Since this test condition had horizontal slats with
relatively large spacing, even when the blinds were deployed, the room still had a certain amount
of daylight available. The average percentage of areas with sufficient daylighting during the day
was 23.35%. The sDA on September 21
st
was 32.12%.

57

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-4 Momentary Illuminance of Test Condition #2
4.1.1.3 Test Condition #3 (D+M)
The third test condition had miniature blinds like the ones typically used as interior blinds. The
width of the slats was 0.03 m, and the spacing was 0.02 m. They had different aspect ratios from
the previous test condition, and the slats were not horizontal, but with 45° tilt angles. At 3 and 4
pm, the blinds were deployed because the incident solar irradiance values at 3 and 4 pm were the
greatest. For the rest of the day, the daylighting of the room was the same (FIG. 4-5). Although
the percentage of area with illuminance larger than 300 lux at 3 and 4 pm was zero, the room was
not completely dark. Daylight was still available for the room. However, it was much darker than
the room using horizontal blinds with larger spacing. The daylight availability of miniature

58
blinds was not as good as horizontal blinds because they blocked more sunlight out of the room.
The average percentage of areas with sufficient daylighting for test condition #3 was 18.23%.
The sDA on the fall equinox was 27.17%.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-5 Momentary Illuminance of Test Condition #3
4.1.1.4 Test Condition #4 (D+O)
The fourth test condition was double-pane glazing with a static overhang. The static overhang
reduced some daylight when there was not enough daylight in the morning. However, in the
afternoon, the overhang efficiently shielded sunlight. The overhang had the same aspect ratio as
the horizontal blinds. It was expected to have the same effect as test condition #2. Whereas at 3

59
and 4 pm, it had less percentage of area with 300 lux than test condition #2, which meant it could
shield more sunlight than the horizontal blinds (FIG. 4-6). In the meantime, it did not block too
much daylight, so the room remained bright. Its daylight performance was better than miniature
blinds. The average percentage of area with a sufficient amount of daylight for this test condition
was 16.88%. The sDA on September 21
st
was 25.52%.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-6 Momentary Illuminance of Test Condition #4
4.1.1.5 Test Condition #5 (E)
This test condition did not have any exterior shading system. It had EC glazing instead of static
double-pane glazing. The EC glazing kept changing its state to shield the sunlight (FIG. 4-7). At

60
7 am, it was in its clear state. From 8 am, its light state was triggered, so the percentage of
illuminance larger than 300 lux was reduced a little bit compared with test condition #1. At noon,
its heavy state was triggered. In the afternoon, it was fully tinted. The amount of natural daylight
in the morning and noon was kept. The uncomfortable sun exposure was largely reduced. In the
evening, it was back to the light state. From the simulation results, it was clear to see that the
daylight went through the window even though the glazing was fully tinted. Therefore, when
there was too much sunlight exposure, EC glazing along was not effective enough to cut the
solar heat out of the building. However, for the daylight availability, EC glazing blocked a
certain amount of sunlight, while still keeping some daylight for the room so it would not be too
dark. The average percentage of areas with illuminance larger than 300 lux on the fall equinox
for test condition #5 was 13.37%. The sDA on the fall equinox was 17.45%.

61

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-7 Momentary Illuminance of Test Condition #5
4.1.1.6 Test Condition #6 (E+H)
The sixth test condition was EC glazing with exterior kinetic horizontal blinds. The kinetic
blinds and EC glazing did not shield sunlight in the morning when it was insufficient (FIG. 4-8).
The kinetic blinds were deployed only at 3 and 4 pm because the solar irradiance on the window
is the highest during that time. The daylighting of the rest of the day was the same as test
condition #5. At 3 and 4 pm, the percentage of area with illuminance larger than 300 lux was
reduced a little, but there was still daylight available. The average percentage of area with
sufficient daylighting for test condition #6 was 11.72%. The sDA on the autumn equinox was
15.71%.

62

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-8 Momentary Illuminance of Test Condition #6
4.1.1.7 Test Condition #7 (E+M)
The seventh test condition had EC glazing and kinetic miniature blinds. Except for 3 and 4 pm,
other hours had the same results as test condition #5 since blinds were not deployed (FIG. 4-9).
However, when miniature blinds were deployed, and EC glazing was switched to the fully tinted
state at the same time, there was no daylight available for the room. Test condition #7’s daylight
performance was not as good as other test conditions. The average percentage of area with
illuminance larger than 300 lux for test condition #7 was 9.55%. The sDA on September 21
st
was
14.06%.

63

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-9 Momentary Illuminance of Test Condition #7
4.1.1.8 Test Condition #8 (E+O)
The last test condition had EC glazing and the static overhang. With the combination of shading
technologies, the working surface area had less daylight availability, but reduced uncomfortable
high illuminance area compared to EC glazing without exterior shades (FIG. 4-10). The
overhang did not block any views, but it shielded more daylight than test condition #6. At 1 and
2 pm, there was very little daylight available. At 3 pm, it provided more daylighting than test
condition #6, while at 4 pm, there was less percentage of area with sufficient daylighting than
test condition #6. Overall, test condition #8 had the least percentage of area with enough
daylighting on the day of the fall equinox, but the room did not have one completely dark

64
moment. However, the average percentage of areas with illuminance larger than 300 lux for the
last test condition was 6.16%, which was the lowest one among all test conditions. The sDA on
the fall equinox was 9.64%.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-10 Momentary Illuminance of Test Condition #8
To summarize the daylight performance of the shoebox model, the eight test conditions had
decreased daylight performance from #1 to #8. Since there was only one small window on the
west façade with a WWR of 30%, the overall daylighting was not sufficient. The shading and
glazing systems reduced the daylight availability.

65
4.1.2 Glare
Glare was a significant indicator of visual comfort and building performance. The table (Table 4-
4) below shows hours of glare situations. The fisheye diagram (FIG. 4-11) facing the window
was created to show the sky luminance (FIG 4-12). “Small Multiples” of sky luminance have
been graphed hourly from 7 am to 6 pm on the fall equinox. The scale (FIG. 4-13) of the unit
was 0 to 3000 candela per square meter (cd/m
2
). The simulation also recorded its Daylight Glare
Probability (DGP), which indicated whether the glare was perceptible or tolerable. DGP was
used as the glare indicator of the experiment and would be used to rank the performance of the
eight test conditions.
Table 4-4 Glare Situations of All Test Conditions
Test Conditions / Glare
(hrs.)
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Imperceptible 4 4 6 4 7 7 9 9
Perceptible 2 2 2 3 2 2 2 1
Disturbing 1 1 1 2 0 0 0 0
Intolerable 5 5 3 3 3 3 1 2

Figure 4-11 Fisheye Scene Figure 4-12 Orientation Figure 4-13 Scale of Sky Luminance

66
4.1.2.1 Test Condition #1 (D)
The first test condition did not have any exterior shades or special glazing, so it had the most
severe impact by glare (FIG. 4-14). The window was on the west façade of the shoebox model,
so the sky luminance was low in the morning, increased in the afternoon, and then decreased in
the evening. The horizon was bright around noon because it reflected the sun at the zenith. In the
afternoon, the sun entered the view, so the sky luminance increased and DGP peaked from 3 to 5
pm. From 7 to 9 am, there was imperceptible glare. From 10 to 11 am, the glare became
perceptible. At noon, it became disturbing. From 1 to 5 pm, the glare had been intolerable. At 6
pm, the sun was about to set, the glare turned imperceptible again. From the simulation results, it
was obvious that dynamic glazing and exterior shading systems were necessary to reduce glare
and improve visual comfort.

67

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-14 Sky Luminance and DGP of Test Condition #1
4.1.2.2 Test Condition #2 (D+H)
The second test condition was static double-pane glazing with exterior horizontal blinds. The
blinds were deployed at 3 and 4 pm (FIG. 4-15). However, glare was still intolerable during that
time. The horizontal louver blinds with relatively large spacing did not shield too much
discomfort glare. Therefore, the glare results of test condition #2 were almost the same as test
condition #1.

68

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-15 Sky Luminance and DGP of Test Condition #2
4.1.2.3 Test Condition #3 (D+M)
The third test condition had static clear glazing but miniature louver blinds with angles of 45°. At
3 and 4 pm, when the blinds were deployed, DGP was reduced from intolerable to imperceptible
(FIG. 4-16). However, there was still disturbing glare at 12 pm, and intolerable glare at 1, 2, and
5 pm. Kinetic miniature blinds could effectively shield glare, but they almost blocked the whole
window. The sky luminance was around 300 cd/m
2
when they were deployed, which was pretty
low. Test condition #3 solved the glare issue when glare was most severe but failed to reduce
less severe but still disturbing glare situations.

69

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-16 Sky Luminance and DGP of Test Condition #3
4.1.2.4 Test Condition #4 (D+O)
The fourth test condition was static double-pane glazing with an overhang. Different from test
condition #3, it failed to solve the most severe glare issue in the afternoon, but effectively
reduced glare when glare was less severe but still bothered occupants around noon time (FIG. 4-
17). From 7 to 9 am, the glare was imperceptible. From 10 am to 12 pm, the glare was
perceptible but not disturbing. At 1 and 2 pm, the glare was disturbing but not intolerable as test
condition #1. From 3 pm to 5 pm, the glare was still intolerable. At 6 pm, the glare became
imperceptible again.

70

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-17 Sky Luminance and DGP of Test Condition #4
4.1.2.5 Test Condition #5 (E)
The fifth test condition was EC glazing without any exterior shades. The EC glazing turned
tinted when incident solar irradiance on the window was high. Therefore, the sky luminance in
the afternoon was low, especially at 3 and 4 pm (FIG. 4-18). However, sky luminance was only
one component of DGP. DGP from 3 pm to 5 pm was reduced from 1 to 0.5 and 0.6, but it was
still intolerable. From 7 to 9 am, glare was imperceptible. At 10 and 11 am, the glare was
perceptible. From 12 to 2 pm, EC glazing changed its tinted states and made the glare
imperceptible instead of disturbing or intolerable. From 3 pm to 5 pm, although the EC glazing

71
was changing from fully tinted and heavy states, glare was still intolerable for occupants. At 6
pm, solar irradiance on the window dropped, so EC glazing turned clear again, and the glare was
still imperceptible.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-18 Sky Luminance and DGP of Test Condition #5
4.1.2.6 Test Condition #6 (E+H)
The sixth test condition had EC glazing and exterior horizontal blinds. Similar to test condition
#2, exterior horizontal louver blinds did not have good performance in reducing discomfort glare
(FIG. 4-19). When the blinds were deployed at 3 and 4 pm, DGP reduced only a little compared
with test condition #5. At 4 pm, DGP reduced from 0.55 to 0.45, and the glare was still

72
intolerable. Therefore, the simulation results of test condition #6 were almost the same as test
condition #5.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-19 Sky Luminance and DGP of Test Condition #6
4.1.2.7 Test Condition #7 (E+M)
The seventh test condition had EC glazing and kinetic miniature blinds. The miniature louver
blinds with small spacings and 45° of slats’ angles almost block all the glare completely when
they were deployed (FIG. 4-20). The sky luminance at 3 and 4 pm was almost 0 cd/m
2
, so the
glare was imperceptible. For the rest of the day, the glare situation was the same as test condition
#5 since only EC glazing was working. Hence, test condition #7 only had intolerable glare at 5

73
pm when blinds were not deployed but tinted glazing was not sufficient to shield the glare.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-20 Sky Luminance and DGP of Test Condition #7
4.1.2.8 Test Condition #8 (E+O)
The last test condition had EC glazing with a static overhang. The overhang shielded the glare to
some extent (FIG. 4-21). Except for 11 am, 4 pm, and 5 pm, other hours all had imperceptible
glare conditions. At 11 am, the glare was perceptible. At 4 and 5 pm, the glare was intolerable.
Compared with test condition #5 which only EC glazing was used, the perceptible glare at 10 am
and intolerable glare at 3 pm were largely improved. The overhang did not effectively block the
glare like miniature blinds, but it reduced the glare while not reducing sky luminance to zero.

74

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-21 Sky Luminance and DGP of Test Condition #8
To summarize, the glazing and shading systems were effective in improving glare performance.
The horizontal blinds did not improve the glare performance, but the miniature blinds and EC
glazing did. Test condition #7 had the best glare performance, and test condition #8 had the
second-best glare performance.
4.1.3 Peak Cooling Load
The peak cooling loads measured the peak mechanical cooling loads on the summer design day,
which was August 21
st
in this experiment. It was an effective indicator of energy performance
because Los Angeles was selected to be the city for the experiment, and it had a cooling-

75
dominated climate. The room had HVAC systems and non-operable windows. This section
records and discusses all the data of cooling loads on the day and when they peaked. The table
below (Table 4-5) shows peak cooling loads of eight test conditions. The unit of peak cooling
load was kilowatt (kW).
Table 4-5 Peak Cooling Loads of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Peak Cooling Load (kW) 2.55 2.06 1.91 2.16 1.79 1.55 1.44 1.71
4.1.3.1 Test Condition #1 (D)
The peak cooling load of test condition #1 (FIG. 4-22) was 2.55 kilowatts (kW). The x-axis
indicated the hours of August 21
st
. The cooling loads peaked at 4 pm. Morning and evening had
a clear drop in cooling loads, and during the night, the cooling load was zero. It should be the
highest value among other test conditions since test condition #1 did not contain any dynamic
glazing or shading to decrease the solar heat of the room.

Figure 4-22 Peak Cooling Load of Test Condition #1
4.1.3.2 Test Condition #2 (D+H)
The second test condition had its cooling load peaked at 4 pm on the summer design day (FIG. 4-
23). The diagram shows the spike disappeared, and a concave trend took its place. This was

76
because the kinetic horizontal blinds reduced the cooling loads when they were deployed in the
afternoon. The rest of the trend seemed to be almost the same as the first test condition.

Figure 4-23 Peak Cooling Load of Test Condition #2
4.1.3.3 Test Condition #3 (D+M)
The third test condition had a more obvious sunken part in the afternoon (FIG. 4-24). Its cooling
load peaked at 1 pm, which was 1.91 kW. Cooling loads at 3 and 4 pm were largely reduced. The
miniature louver blinds were very effective in reducing cooling loads.

Figure 4-24 Peak Cooling Load of Test Condition #3
4.1.3.4 Test Condition #4 (D+O)
The fourth test condition had its peak cooling load of 2.16 kW at 5 pm (FIG. 4-25). A slightly
less sharp spike could still be seen from the trend. The peak cooling load was only decreased a

77
little compared to test condition #1, and it was higher than test conditions #2 and #3.

Figure 4-25 Peak Cooling Load of Test Condition #4
4.1.3.5 Test Condition #5 (E)
The fifth test condition was EC glazing that kept changing states as the incident solar irradiance
changed on the window (FIG. 4-26). There was no spike on the trend line and there was a small
concave part. The EC glazing reduced a large amount of solar heat in the afternoon and a little
around noon. The cooling load peaked at 5 pm.

Figure 4-26 Peak Cooling Load of Test Condition #5
4.1.3.6 Test Condition #6 (E+H)
The sixth test condition was EC glazing with kinetic horizontal blinds (FIG. 4-27). The addition
of kinetic louver blinds reduced the peak cooling load to 1.55 kW at 5 pm. The trend line of test

78
condition #6 was flatter than test condition #2 which kinetic horizontal blinds were used alone
but could still see the peak time in the afternoon. The value reduced a lot, which meant that
kinetic horizontal blinds were an effective addition to the EC glazing of reducing cooling energy.

Figure 4-27 Peak Cooling Load of Test Condition #6
4.1.3.7 Test Condition #7 (E+M)
Test condition #7 had a flatter trend line because the miniature blinds were more effective in
blocking out the solar heat at the peak moment (FIG. 4-28). The cooling load peaked at 2 pm was
1.44 kW. Miniature blinds with 45° of slats shielded more solar heat than horizontal blinds with
larger spacing. Therefore, EC glazing with kinetic miniature blinds had the peak cooling load
lower than test conditions #5 and #6.

Figure 4-28 Peak Cooling Load of Test Condition #7

79
4.1.3.8 Test Condition #8 (E+O)
The last test condition was EC glazing working with the static overhang (FIG. 4-29). The trend
line did not have a concave part because the overhang was always there to reduce the solar heat.
The peak cooling load was at 5 pm. It was 1.71 kW, which was a little lower than test condition
#5. This made the data valid since the static overhang was an addition to the EC glazing. Test
condition #8 had a higher peak cooling load than test conditions #6 and #7, but it was also
considered effective in reducing mechanical cooling loads.

Figure 4-29 Peak Cooling Load of Test Condition #8
To sum up, test condition #7 had the lowest peak cooling loads. Test condition #6 had the
second-lowest peak cooling load. All glazing and shading strategies had good performance
reducing cooling loads and annual cooling consumption.
4.1.4 EUI
Energy Use Intensity (EUI) was another indicator to assess energy performance. It took more
factors into account. While the dynamic glazing and shading systems shield solar heat and
reduce cooling loads, they would also increase heating loads even if the building was located in a
city with a warm climate. EUI would consider the trade-offs of cooling, heating, and electric
lighting EUI. While the equipment energy use was the same for all test conditions, the electric

80
lighting use changed with daylight performance. The percentage of area with illuminance larger
than 300 lux was considered bright enough so that artificial lighting was not necessary. Hence,
this amount of electric lighting would be reduced from the EUI value. The test condition with the
lowest value of EUI had the greatest efficiency in reducing energy use. The table below (Table
4-6) records the EUI breakdowns of all test conditions.
Table 4-6 EUI Breakdowns of All Test Conditions
Test Conditions \
EUI (kWh/m
2
)
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Heating 1.74 1.79 1.85 2.32 1.91 2.43 3.30 2.20
Cooling 16.44 15.86 14.58 11.34 9.72 9.43 7.29 9.03
Interior Lighting 21.38 21.65 23.09 23.47 24.46 24.93 25.54 26.50
Interior Equipment 34.32 34.32 34.32 34.32 34.32 34.32 34.32 34.32
Original EUI 80.73 80.21 78.99 76.22 74.19 74.36 73.09 73.79
Lighting Deducted 6.86 6.59 5.15 4.77 3.78 3.31 2.7 1.74
New Interior Lighting 21.38 21.65 23.09 23.47 24.46 24.93 25.54 26.50
New EUI 73.86 73.61 73.84 71.45 70.41 71.05 70.39 72.05
4.1.4.1 Test Condition #1 (D)
Test condition #1 had static double-pane glazing only. Without dynamic glazing and exterior
shading systems, it accepted the greatest amount of solar heat among all test conditions.
Therefore, its annual heating energy use should be the lowest, but its cooling should be the
highest (Table 4-7). While the equipment had the same value for all test conditions, the interior
lighting consumed the least amount of energy because test condition #1 also had the greatest
amount of daylight availability. 6.86 kWh/m
2
of energy use was subtracted from the original EUI
value of 80.73 kWh/m
2
. The final EUI value of the first test condition was 73.86 kWh/m
2
.
Table 4-7 EUI Breakdowns of Test Condition #1
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
1.74 16.44 28.24 34.32 80.73 6.86 73.86

81
4.1.4.2 Test Condition #2 (D+H)
The second test condition had its annual heating energy use increased and cooling decreased.
The original EUI was a little lower than test condition #1 (Table 4-8), which was 80.21 kWh/m
2
.
The amount of interior lighting reduced was 6.59 kWh/m
2
, and the overall EUI was 73.61
kWh/m
2
. The results were very close to but slightly lower than test condition #1.
Table 4-8 EUI Breakdowns of Test Condition #2
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
1.79 15.86 28.24 34.32 80.21 6.59 73.61
4.1.4.3 Test Condition #3 (D+M)
The third test condition had an increased value of annual heating and a decreased value of annual
cooling as well (Table 4-9). Its original EUI was 78.99 kWh/m
2
, which was lower than test
condition #3. However, after the amount of 5.15 annual interior lighting was subtracted from the
original EUI, the new EUI value of 73.84 kWh/m
2
was a little higher than the previous test
condition. This meant the miniature blinds of test condition #3 need more electric lighting to
compensate for the sunlight it had blocked.
Table 4-9 EUI Breakdowns of Test Condition #3
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
1.85 14.58 28.24 34.32 78.99 5.15 73.84
4.1.4.4 Test Condition #4 (D+O)
The static overhang shielded solar heat all day long, so it was more effective in reducing annual
cooling energy than peak cooling energy. The annual cooling consumption of test condition #4
was 11.34 kWh/m
2
(Table 4-10), which was much lower than test conditions #1, #2, and #3. The

82
annual heating energy use of test condition #4 was higher than test condition #3, and its annual
cooling energy use was lower than test condition #3. The original EUI was 76.22 kWh/m
2
. The
amount of 4.77 kWh/m
2
interior lighting energy use was removed from the original EUI, and the
new EUI was 71.45 kWh/m
2
. All these values were lower than the previous test condition.
Table 4-10 EUI Breakdowns of Test Condition #4
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
2.32 11.34 28.24 34.32 76.22 4.77 71.45
4.1.4.5 Test Condition #5 (E)
Test condition #5 had both annual heating and cooling energy use lower than the previous test
condition (Table 4-11). It meant EC glazing had better performance in saving both mechanical
heating and cooling energy than static glazing with a static overhang. The original EUI was
74.19 kWh/m
2
. The artificial lighting use that could be reduced from the EUI was 3.78 kWh/m
2
.
The new EUI after the reduction was 70.41 kWh/m
2
, which was lower than all test conditions
with static double-pane glazing.
Table 4-11 EUI Breakdowns of Test Condition #5
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
1.91 9.72 28.24 34.32 74.19 3.78 70.41
4.1.4.6 Test Condition #6 (E+H)
Test condition #6 had EC glazing and kinetic horizontal louver blinds with large spacings of
slats. The heating EUI of the year increased a lot from test condition #5, but the cooling use
decreased even more (Table 4-12). The original EUI of 74.36 kWh/m
2
was lower than the
previous one. The amount of artificial lighting that could be reduced was 3.31 kWh/m
2
, which

83
was also lower. The new EUI value was 71.05 kWh/m
2
. The EUI improved from the previous
again.
Table 4-12 EUI Breakdowns of Test Condition #6
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
2.43 9.43 28.24 34.32 74.36 3.31 71.05
4.1.4.7 Test Condition #7 (E+M)
Similar to test condition #3, test condition #7 had a lower value of original EUI 73.09 kWh/m
2

compared to the previous test condition that had the same glazing type but a different louver type
(Table 4-13). However, after the sufficient daylight compensation of 2.70 kWh/m
2
was reduced
from the interior lighting energy use, the new EUI value of 70.39 kWh/m
2
was slightly higher
than test condition #6. Too much energy from electric lighting was increased compared to the
cooling energy reduced.
Table 4-13 EUI Breakdowns of Test Condition #7
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
3.30 7.29 28.24 34.32 73.09 2.70 70.39
4.1.4.8 Test Condition #8 (E+O)
The last test condition had a smaller value of heating use and a greater value of cooling use
compared to the previous test condition (Table 4-14). Nevertheless, the heating use increased,
and the cooling use decreased compared to test condition #5 that EC glazing was used alone.
However, its electric lighting use that could be compensated by sufficient daylighting was the
smallest among all test conditions. Therefore, although its original EUI was 73.79 kWh/m
2
,
smaller than test condition #5 when the interior lighting use 1.74 kWh/m
2
was reduced, the new

84
EUI was 72.05 kWh/m
2
, higher than all other test conditions with EC glazing.
Table 4-14 EUI Breakdowns of Test Condition #8
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
2.20 9.03 28.24 34.32 73.79 1.74 72.05
In conclusion, EUI considered heating and cooling consumption and daylight availability.
Therefore, it showed an overall performance. The one that had the lowest EUI value was test
condition #7 (EC glazing + kinetic 45° miniature blinds) and the second-lowest one was test
condition #5 (EC glazing).
4.1.5 Adaptive Thermal Comfort
Both experiments #1 and #2 were in air-conditioned and non-naturally ventilated buildings.
However, the situation of the power outage accident was considered. The dynamic shading and
glazing were active strategies, and they relied on electricity. Therefore, only the situations of
static double-pane glazing and static overhang were considered for adaptive thermal comfort
when the HVAC system was not available, but the windows were still not operable.
Performance of thermal comfort contained Adaptive Thermal Comfort and the extreme positive
indoor temperatures deviation from the neutral temperature zone of a year. This section
discussed the percentage of time that fell into the neutral temperature range during a year. The
diagram (FIG. 4-30) shows the neutral temperatures for each month. In summer, the neutral
temperature is 24.01 °C. In winter, the neutral temperature is 22.06 °C. The conditions with
neutral temperatures are comfortable times. The tables record the data of percentages of thermal
conditions.

85

Figure 4-30 Neutral Temperatures of the Year
4.1.5.1 Test Condition #1 (D)
The red part in the diagram shows the overheated area that the temperature was 2.5 °C higher
than the neutral temperature of the month (FIG. 4-31). The blue part shows the cold conditions
that the temperature was 2.5 °C lower than the neutral temperature of the month. The yellow part
shows the comfortable conditions when the temperature was within the 2.5 °C offsets.

Figure 4-31 Adaptive Thermal Comfort of Test Condition #1
This function is regulated by ASHRAE 55 (ANSI/ASHRAE 2017). When the room had no
HVAC, there was 71.83% overheated time and 27.28% comfortable time. The highest indoor
temperature was 36.49 °C (Table 4-15). The highest positive deviation from the neutral
temperature was 12.88 °C (FIG. 4-32). The overall indoor temperature was uncomfortable but
not fatal to human beings.

86
Table 4-15 Thermal Information of Test Condition #1
% Hot % Neutral
(Comfortable)
% Cold Lowest
Temperature
°C
Highest
Temperature
°C
Lowest
Deviation
°C
Highest
Deviation
°C
71.83 27.28 0.89 17.98 36.49 -4.08 12.88

Figure 4-32 Deviation from Neutral Temperatures of Test Condition #1
4.1.5.2 Test Condition #4 (D+O)
With the static overhang, test condition #4 had 61.23% of the overheated time and 37.33% of the
comfortable time when the room had no HVAC (Table 4-16). The highest indoor temperature
was 33.59 °C. The diagram shows the distribution of adaptive thermal comfort (FIG. 4-33). The
highest positive deviation from the neutral temperature was 9.65 °C (FIG. 4-34). The passive
shading strategy improved the thermal situation a little bit. However, the room was still
uncomfortable. The next section discussed the situation with air-conditioning.
Table 4-16 Thermal Information of Test Condition #4
% Hot % Neutral
(Comfortable)
% Cold Lowest
Temperature
°C
Highest
Temperature
°C
Lowest
Deviation
°C
Highest
Deviation
°C
61.23 37.33 1.44 17.56 33.59 -4.50 9.65

Figure 4-33 Adaptive Thermal Comfort of Test Condition #4

87

Figure 4-34 Deviation from Neutral Temperatures of Test Condition #4
4.1.6 PMV Thermal Comfort
Performance of thermal comfort contained Predicted Mean Vote (PMV) thermal comfort
conditions and Predicted Percentage of Dissatisfied (PPD) of a year. The two metrics were used
for air-conditioned rooms. As described in Chapter 3, occupants voted for thermal conditions
from -3 to +3. 0 stood for neutral temperatures, and the range of -0.5 to +0.5 stood for neutral
zone. Anything above +0.5 was too hot, and anything below -0.5 was too cold. The percentage of
the neutral zone was the first indicator of thermal performance.
For the large office building type, there were different HVAC heating and cooling setpoints for
occupied and unoccupied hours. From 6 am to 9 pm, the heating setpoint was 21 °C, and the
cooling setpoint was 24 °C. For the rest of the hours, the heating setpoint was 15.6 °C, and the
cooling setpoint was 26.7 °C.
Both PMV and PPD results showed uncomfortable cold conditions in the winter, and those were
caused by HVAC systems’ heating setpoints. Hence, cold conditions would be excluded from
PMV and PPD analyses and scores. The table (Table 4-17) records the data of percentages of
thermal conditions. Since different glazing and shading strategies were used to reduce
overheated conditions, a percentage of overheated time would be used in the scoring system. If
more than one test condition had no overheated time, they all had the first rank in the PMV

88
thermal performance and earned eight points.
Table 4-17 Thermal Conditions of All Test Conditions
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6 E+H #7
E+M
#8 E+O
% Hot 0.64 0 0 0 0 0 0 0
%
Comfortable
72.07 71.50 70.59 67.75 66.99 62.77 56.13 64.62
% Cold 27.29 28.50 29.41 32.25 33.01 37.23 43.87 35.38
4.1.6.1 Test Condition #1 (D)
The orange part in the diagram shows the overheated area that the thermal comfort was larger
than +0.5 from the -3 to +3 scale rated by occupants. The light blue part shows the cold
conditions that the thermal comfort was lower than -0.5 (FIG. 4-35). The yellow part shows the
comfortable conditions when the temperature was between -0.5 and +0.5. The humid and dry
parts were not considered in the diagram.

Figure 4-35 Thermal Condition of Test Condition #1
Since the shoebox model only had one window on the west façade, and the HVAC system was
available, the overheated time of test condition #1 was little. The hot percentage was 0.64%, the
comfortable percentage was 72.07%, and the cold percentage was 27.29% (Table 4-18). The
diagram shows that the overheated time happened in the afternoon of summer, and the cold time
happened in the morning and evening of the winter.

89
Table 4-18 Thermal Condition of Test Condition #1
% Too Hot % Comfortable % Too Cold
0.64 72.07 27.29
4.1.6.2 Test Condition #2 (D+H)
When the room was air-conditioned, test condition #2 had no overheated time (FIG. 4-36). The
cold time was 28.50%, which was slighter greater than test condition #1 (Table 4-19). The
comfortable percentage was 71.50%, which was greater than the static glazing without exterior
shades.

Figure 4-36 Thermal Condition of Test Condition #2

Table 4-19 Thermal Condition of Test Condition #2
% Too Hot % Comfortable % Too Cold
0 71.50 28.50
4.1.6.3 Test Condition #3 (D+M)
When the room was air-conditioned, the third test condition had no overheated time (FIG. 37). It
also had less comfortable time of 70.59% and more cold time of 29.41% (Table 4-20).

Figure 4-37 Thermal Condition of Test Condition #3

90

Table 4-20 Thermal Condition of Test Condition #3
% Too Hot % Comfortable % Too Cold
0 70.59 29.41
4.1.6.4 Test Condition #4 (D+O)
Test condition #4 had no hot time (FIG. 4-38). The comfortable time dropped a lot to 67.75%
(Table 4-21). The cold time increased to 32.25%.

Figure 4-38 Thermal Condition of Test Condition #4

Table 4-21 Thermal Condition of Test Condition #4
% Too Hot % Comfortable % Too Cold
0 67.75 32.25
4.1.6.5 Test Condition #5 (E)
The comfortable time kept dropping because the test conditions had more effective shading
strategies (FIG. 4-39). The cold time of test condition #5 increased to 33.01% (Table 4-22). The
comfortable decreased to 66.99%.

Figure 4-39 Thermal Condition of Test Condition #5

91
Table 4-22 Thermal Condition of Test Condition #5
% Too Hot % Comfortable % Too Cold
0 66.99 33.01
4.1.6.6 Test Condition #6 (E+H)
The sixth test condition had no hot time during the year when air-conditioned (FIG. 4-40). Its
comfortable time was 62.77%, which was a lot smaller than the previous one because the kinetic
horizontal blinds created more cold time in addition to EC glazing (Table 4-23). The cold time
was 37.23%.

Figure 4-40 Thermal Condition of Test Condition #6

Table 4-23 Thermal Condition of Test Condition #6
% Too Hot % Comfortable % Too Cold
0 62.77 37.23
4.1.6.7 Test Condition #7 (E+M)
The air-conditioned seventh test condition had the lowest percentage of comfortable time (FIG.
4-41), which was 56.13% (Table 4-23). However, it did not mean the combination of EC glazing
and kinetic miniature blinds had bad performance in thermal comfort. The shoebox model with
one window did not have the same level of solar heat as a standard office building. Therefore,
test condition #7 had the highest value of cold percentage of the time, which was 43.87%.

92

Figure 4-41 Thermal Condition of Test Condition #7

Table 4-24 Thermal Condition of Test Condition #7
% Too Hot % Comfortable % Too Cold
0 56.13 43.87
4.1.6.8 Test Condition #8 (E+O)
The last test condition had a higher percentage of comfortable time compared to test conditions
#6 and #7 when it had the HVAC system (FIG. 4-42), which was 64.62% (Table 4-25). The cold
time was less than the previous two conditions, which was 35.38%.

Figure 4-42 Thermal Condition of Test Condition #8

Table 4-25 Thermal Condition of Test Condition #8
% Too Hot % Comfortable % Too Cold
0 64.62 35.38
In summation, test condition #1 had the highest percentage of comfortable time for the air-
conditioned shoebox model room. However, it was the only test condition that had overheated
time. All test conditions had more percentage of cold time than hot time. This was because the
overheated time could not be drastically influenced by the shading and glazing systems since the

93
shoebox model only had a small window on the west façade. Test condition #5 had the lowest
percentage of the comfortable time, but test conditions #2 to #8 all had the best PMV thermal
performance due to 0% overheated time.
4.1.7 PPD
The other indicator of thermal performance was the greatest Predicted Percentage Dissatisfied
(PPD) value of the year. When the previous indicator considered the percentage of the
comfortable time, this considered the level of uncomfortable time. The worst overheated level
could be ameliorated by shading or glazing strategies. The PMV Thermal Comfort could omit
situations that rooms with similar percentages of comfortable time could have different levels of
uncomfortable time. The dissatisfied time was when PMV was outside the range of -0.5 to +0.5.
The range of PPD was from 5% to 100%. The lowest possible PPD value was 5%, and the range
from 5% to 10% was comfortable, but values lower than 20% were considered efficient by
ASHRAE 55 as described in Chapter 3. Anything from 20% to 100% was disturbing. Therefore,
the scale used for PPD was from 5% to 20% (FIG. 4-43). Only air-conditioned rooms were
simulated for all test conditions, and only PPD values for the overheated conditions were
considered for the evaluation. Since the PPD values for cold conditions were caused by the 15.6
°C HVAC heating setpoint during unoccupied hours in the winter, they were not caused by the
shading or glazing strategies and would not influence occupants. The table (Table 4-26) records
the PPD values for cold and overheated conditions and the highest and lowest indoor
temperatures of a year.

94

Figure 4-43 Scale of PPD

Table 4-26 PPD and Indoor Temperatures of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Lowest °C 16.8 16.8 16.8 16.64 16.85 16.56 16.04 16.63
Highest °C 25.88 25.59 25.55 25.5 25.08 24.96 24.78 24.95
PPD % for the Hottest Time 11.93 10.00 9.68 9.77 7.57 6.94 6.62 7.11
PPD % for the Coldest Time 84.74 87.37 84.74 86.87 82.97 85.29 89.21 84.37
4.1.7.1 Test Condition #1 (D)
The color scale represented uncomfortable levels. The greater the percentage, the redder the
color (FIG. 4-44). Therefore, the red color represented both overheated and cold conditions. For
this diagram, the red color happened most in the mornings and evenings of the winter, meaning it
was uncomfortable because it was too cold.

Figure 4-44 PPD of Test Condition #1

The highest temperature of test condition #1 was 25.88 °C (Table 4-27), which was not high. The
highest PPD value for the cold conditions was 84.74%, which happened during the unoccupied
hours in winter. The highest PPD value for the overheated conditions was 11.93% happened in

95
one afternoon of September.
Table 4-27 PPD and Temperatures of Test Condition #1
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.8 25.88 11.93 84.74
4.1.7.2 Test Condition #2 (D+H)
Test condition #2 had a lower indoor temperature, which was 25.59 °C (Table 4-28). The highest
PPD for the overheated conditions was 10.00%, which was a little lower than test condition #1,
meaning this shading strategy decreased uncomfortable levels (FIG. 4-45).

Figure 4-45 PPD of Test Condition #2

Table 4-28 PPD and Temperatures of Test Condition #2
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.8 25.59 10.00 87.37
4.1.7.3 Test Condition #3 (D+M)
Test condition #3 had 25.55 °C of the highest indoor temperature (Table 4-29). The PPD value
for the hottest time was 9.68%. The diagram shows PPD distribution over the year (FIG. 4-46).

96

Figure 4-46 PPD of Test Condition #3

Table 4-29 PPD and Temperatures of Test Condition #3
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.8 25.55 9.68 84.74
4.1.7.4 Test Condition #4 (D+O)
Test condition #4 had 25.50 °C of the highest indoor temperature (Table 4-30). The highest PPD
for the overheated condition was 9.77%. It was higher than test condition #3. The PPD
distribution did not change too much (FIG. 4-47).

Figure 4-47 PPD of Test Condition #4

Table 4-30 PPD and Temperatures of Test Condition #4
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.64 25.50 9.77 86.87
4.1.7.5 Test Condition #5 (E)
Test condition #5 had 25.08 °C of the highest indoor temperature (Table 4-31). The PPD value
for the worst overheated time was 7.57%. It was lower than the previous test conditions. There

97
were more blue parts in the diagram in the summer, meaning the summer had a less
uncomfortable time (FIG. 4-48).

Figure 4-48 PPD of Test Condition #5

Table 4-31 PPD and Temperatures of Test Condition #5
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.85 25.08 7.57 82.97
4.1.7.6 Test Condition #6 (E+H)
The highest indoor temperature of test condition #6 was 24.96 °C (Table 4-32). The PPD value
for the most overheated time was 6.94%. The PPD distribution was almost the same as the
previous one (FIG. 4-32).

Figure 4-49 PPD of Test Condition #6

Table 4-32 PPD and Temperatures of Test Condition #6
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.56 24.96 6.94 85.29

98
4.1.7.7 Test Condition #7 (E+M)
The highest indoor temperature of test condition #7 was 24.78 °C (Table 4-33). The PPD value
for the hottest time was 6.62%, which was the lowest value among all test conditions (FIG. 4-
50). Since the value was very close to 5.00%, the thermal performance during the summer time
in the occupied hours was almost ideal.

Figure 4-50 PPD of Test Condition #7

Table 4-33 PPD and Temperatures of Test Condition #7
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.04 24.78 6.62 89.21
4.1.7.8 Test Condition #8 (E+O)
The highest indoor temperature of test condition #8 was 24.96 °C (Table 4-34). The highest PPD
value for the overheated conditions was 7.11%. There were more blue parts in April and
November compared to the previous test condition (FIG. 4-51).

Figure 4-51 PPD of Test Condition #8

99
Table 4-34 PPD and Temperatures of Test Condition #8
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.67 24.96 7.11 84.37
All in all, test condition #7 had the lowest value of the PPD value for the worst overheated
conditions. However, the PPD value for the overheated conditions was low for all eight test
conditions, meaning the overheated conditions in the shoebox model were all acceptable when
the air-conditioning was available. Then the shading and glazing systems reduced the PPD value
and made the thermal conditions more comfortable.

100
4.2 Experiment #2 - The Entrada Project
The southwest corner on the third level of the Entrada project was modeled for the experiment
(FIG. 4-52). The model had glass facades on the south and west sides, which were non-operable
windows. There were also mullions on the glass façade and columns inside the office. The
furniture was not included in the simulations. Due to large glass areas, the simulation results
varied a lot from the shoebox model. The two images were the southwest isometric view and the
northeast isometric view of the office model. The two red spots on the facades were sensors used
to detect incident solar irradiance. The arrows indicated the orientation of the sensors.

Figure 4-52 Southwest and Northeast Isometric Views of Entrada Office Corner
4.2.1 Daylighting
The table (Table 4-35) shows the glazing and shading systems for the Entrada project when the
test conditions had dynamic strategies. The south and west façades received different solar
irradiance, so they had different states of shading and glazing systems.

Sensors

101
Table 4-35 States of Dynamic Systems of West and South Facades on 9/21
Time / Façade 7 am 8 am 9 am 10 am 11 am 12 pm
West Clear Clear Clear Clear Light Light
South Light Full Full &
Blinds
Full &
Blinds
Full &
Blinds
Full &
Blinds
Time / Façade 1 pm 2 pm 3 pm 4 pm 5 pm 6 pm
West Light Light Light Light Full &
Blinds
Clear
South Full Heavy Light Clear Clear Clear
The daylighting performance of the Entrada project was expected to be much better than the
shoebox model because of the increasing amount of window areas. Even for the test conditions
with dynamic glazing and shading systems, there should be a lot more daylight availability. The
table below records the average percentage of areas receiving sufficient daylighting on the fall
equinox and the sDA on the same day (Table 4-36). The image (FIG. 4-53) shows the scale of
the illuminance of lux.
Table 4-36 Daylighting Data of All Test Conditions

Figure 4-53 Scale Momentary Illuminance
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% ≥ 300 lux
on 9/21
62.89% 61.13% 41.77% 46.31% 27.24% 23.35% 22.10% 10.27%
sDA300,50%
on 9/21
70.36% 69.32% 65.03% 53.86% 33.56% 31.82% 31.52% 16.02%

102
4.2.1.1 Test Condition #1 (D)
There was a large amount of daylight available for the first test condition (FIG. 4-54). The
average percentage of areas of illuminance larger than 300 lux on the fall equinox was 62.89%.
For most of the hours, there was more than 60% of working areas received sufficient daylight.
The sDA on September 21
st
was 70.36%.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-54 Momentary Illuminance of Test Condition #1

103
4.2.1.2 Test Condition #2 (D+H)
The louver blinds on the south and west facades were not deployed at the same time due to the
variance of incident solar irradiance (FIG. 4-55). From 9 am to 12 pm, kinetic horizontal blinds
were deployed on the south façade. At 5 pm, the blinds were deployed on the west façade.
However, the west façade still got a large amount of sunlight at 5 pm. For the whole day, the
average percentage of areas receiving sufficient daylight was 61.13%, which was almost the
same as test condition #1. The horizontal louver blinds with large spacings did not affect
daylighting too much. The sDA on the fall equinox was 69.32%.

104

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-55 Momentary Illuminance of Test Condition #2
4.2.1.3 Test Condition #3 (D+M)
The third test condition showed very different simulation results compared to the previous ones
(FIG. 4-56). The miniature louver blinds with angles of 45° blocked a lot of daylight when
deployed. When the south façade had the blinds deployed, there was almost no daylight from the
south. The west façade had the same situation. The percentage of working surface areas with

105
sufficient daylight was 41.77% on the fall equinox. The sDA on the fall equinox was 65.03%.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-56 Momentary Illuminance of Test Condition #3
4.2.1.4 Test Condition #4 (D+O)
Test condition #4 had a static overhang for the entire day (FIG. 4-57). Most of the hours have
around 50% of areas receiving sufficient daylight. The average percentage of areas having
illuminance larger than 300 lux was 46.31%. This value was higher than test condition #3 and

106
lower than test condition #2 and showed an overall good daylight performance. The sDA on the
fall equinox was 53.86%.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-57 Momentary Illuminance of Test Condition #4
4.2.1.5 Test Condition #5 (E)
Test condition #5 had dynamic tinted glazing to shield the sunlight according to incident solar
irradiances on the two facades (FIG. 4-58). The EC glazing started changing states from 7 am,
and the impacts were obvious. At 11 pm, only 17% of areas received sufficient daylight. In the

107
afternoon, the EC glazing of both south and west facades shielded daylight. At 6 pm, the EC
glazing turned to clear again. The average percentage with areas receiving more than 300 lux of
illuminance was 27.24%, which still indicated an acceptable amount of daylight. The sDA on the
fall equinox was 31.42%.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-58 Momentary Illuminance of Test Condition #5

108
4.2.1.6 Test Condition #6 (E+H)
Test condition #6 had deployed horizontal louver blinds from 9 am to 12 pm on the south façade,
and at 5 pm on the west façade in addition to the EC glazing (FIG. 4-59). The daylight
performance of those hours seemed to be only affected a little. However, the average percentage
of areas with sufficient daylight availability on that day was 23.35%, which was much lower
than the previous test condition. The sDA on the fall equinox was 30.65%.

109

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-59 Momentary Illuminance of Test Condition #6
4.2.1.7 Test Condition #7 (E+M)
Test condition #7 had the second-lowest average percentage of areas receiving daylight larger
than 300 lux, which was 22.10% (FIG. 4-60). The miniature louver blinds blocked a lot of
daylight in addition to the application of EC glazing. At 11 am, there was only 4% of working
surface areas had sufficient amounts of daylight. Artificial lighting would be necessary at that

110
time. The sDA on the fall equinox was the lowest among all, which was 29.39%.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-60 Momentary Illuminance of Test Condition #7
4.2.1.8 Test Condition #8 (E+O)
The last test condition had an overhang shielding the sunlight all day long and decreased daylight
availability in addition to the EC glazing (FIG. 4-61). At 11 am and 12 pm, there were less than
1% of the areas received illuminance of more than 300 lux. The average percentage of areas with

111
sufficient daylight was 10.27%. The value was the lowest among all test conditions. The sDA on
September 21
st
was 14.27%, which was the lowest value as well.

7 am 8 am 9 am

10 am 11 am 12 pm

1 pm 2 pm 3 pm

4 pm 5 pm 6 pm
Figure 4-61 Momentary Illuminance of Test Condition #8
To sum up, all test conditions could receive sufficient daylighting, while test conditions #8 and
#7 had the least amount of daylight availability. All other test conditions had sDA value of the
fall equinox day larger than 30%.

112
4.2.2 Glare
The position of a seat was chosen (FIG. 4-62), and the direction was looking toward the west,
which was where the computer was facing. This position had both south and west façade in
occupants’ visions, so the glare impact could be properly estimated. A different color palette was
chosen for the Entrada project because it was more complicated than the shoebox model. The
former color palette could not show the mullions and columns of the Entrada project clearly.
Also, the peak point of the sky luminance was shown in the fisheye diagrams. The scale of the
sky luminance shown in the fisheye diagrams was from 0 to 3,000 cd/m
2
(FIG. 4-63). The metric
Daylight Glare Probability (DGP) considered sky luminance, illuminance, and other factors as
described in Chapter 2. The sky luminance map did not directly reflect the value of DGP, but
they were correlated. DGP indicated whether the glare had negative impacts on occupants. The
table below shows the situations of glare situations (Table 4-37).

Figure 4-62 Orientation

113

Figure 4-63 Scale of Sky Luminance

Table 4-37 Glare Situations of All Test Conditions
Test Conditions \
Glare (hrs.)
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Imperceptible 2 6 7 11 11 11 12 11
Perceptible 7 5 5 0 0 0 0 0
Disturbing 2 0 0 0 0 0 0 0
Intolerable 1 1 0 1 1 1 0 1
4.2.2.1 Test Condition #1 (D)
Test condition #1 did not have any shading systems or dynamic glazing, so it showed the most
severe impacts of glare on the office (FIG. 4-64). There was perceptible glare from 8 am to 9 am.
The peak points sometimes were on the reflection of the columns, south façade, or west façade.
10 am and 11 am had disturbing glare from the south façade, which was the left of the view.
From 12 pm to 4 pm, there was perceptible glare moving from the south façade to the west
façade. At 5 pm, the glare was intolerable, and the peak point had a sky luminance of
238,070,000 cd/m
2
, while the peak points of the rest of the time were around 5,000 cd/m
2
only.

114

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-64 Sky Luminance and DGP of Test Condition #1
4.2.2.2 Test Condition #2 (D+H)
From 9 am to 12 pm, horizontal blinds with large spacings between slats on the south façade
were deployed, so perceptible glare from test condition #1 was reduced to imperceptible glare
(FIG. 4-65). At 5 pm, horizontal blinds on the west façade were deployed, but they did not
effectively block the glare. The glare at 5 pm was still intolerable, and the peak point still had a
sky luminance of 238,070,000 cd/m
2
. At 8 am, and from 1 pm to 4 pm, the glare was perceptible
because blinds were not deployed.

115

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-65 Sky Luminance and DGP of Test Condition #2
4.2.2.3 Test Condition #3 (D+M)
From 9 am to 12 pm, miniature blinds with 45° of slats were deployed on the south façade (FIG.
4-66). The glare was imperceptible, and the peak sky luminance was about 3,000 cd/m
2
, while it
was around 5,000 cd/m
2
without blinds. At 5 pm, the blinds on the west façade were deployed,
and the glare was reduced from intolerable to imperceptible. The peak point of sky luminance
was decreased to 7,804 cd/m
2
, which was still a little higher than other hours. No disturbing or
intolerable glare happened to test condition #3, so it was effective in solving glare issues.

116

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-66 Sky Luminance and DGP of Test Condition #3
4.2.2.4 Test Condition #4 (D+O)
The static overhang was fixed, and it reduced some glare in the morning and at noon (FIG. 4-67).
From 7 am to 4 pm, the office only had imperceptible glare. At 5 pm, the glare was intolerable,
and its peak sky luminance was not reduced compared to test condition #1.

117

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-67 Sky Luminance and DGP of Test Condition #4
4.2.2.5 Test Condition #5 (E)
The EC glazing of the south façade started changing to the light state since 7 am (FIG. 4-68).
From 7 am to 4 pm, glare was blocked by tinted glazing on either south or west facades, so it was
imperceptible. The darkest moment was at 12 pm, while the west façade had a light state, and the
south façade was in its fully tinted state. The peak sky luminance was 1,508 cd/m
2
only. At 5 pm,
although the EC glazing of the west façade was at its fully tinted state, it still failed to shield the
glare. DGP was 1, and the glare at 5 pm was intolerable. The peak sky luminance was 7,213,700
cd/m
2
, which was reduced from 238,070,000 cd/m
2
when no EC glazing was applied, but it was

118
still not effective enough.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-68 Sky Luminance and DGP of Test Condition #5
4.2.2.6 Test Condition #6 (E+H)
The kinetic horizontal blinds along with fully tinted glazing decreased the glare all day except
for 5 pm (FIG. 4-69). The combination was not effective enough to solve the intolerable glare
issue. The peak sky luminance was the same as test condition #5.

119

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-69 Sky Luminance and DGP of Test Condition #6
4.2.2.7 Test Condition #7 (E+M)
The miniature blinds were deployed from 9 am to 12 pm on the south façade, and at 5 pm on the
west façade (FIG. 4-70). When the blinds were not deployed, the EC glazing kept the glare
imperceptible for other hours. For the entire day, the glare situation was always imperceptible.
At 5 pm, when the glare was the strongest for other test conditions, the miniature blinds
successfully blocked the glare. The peak sky luminance at 5 pm was 2,636 cd/m
2
, which fell on
the south façade instead of the west façade. The lowest peak sky luminance at 11 am was 1,202
cd/m
2
, which was a little lower than test condition #5. Test condition #7 had the best

120
performance in solving glare issues among all test conditions.

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-70 Sky Luminance and DGP of Test Condition #7
4.2.2.8 Test Condition #8 (E+O)
Similar to test condition #6, test condition #8 kept the glare imperceptible all day except for one
hour (FIG. 4-71). At 5 pm, the overhang plus the fully tinted glazing on the west façade were not
sufficient to reduce the glare from the sunset. The peak sky luminance was the same as test
condition #5.

121

7 am 8 am 9 am 10 am

11 am 12 pm 1 pm 2 pm

3 pm 4 pm 5 pm 6 pm
Figure 4-71 Sky Luminance and DGP of Test Condition #8
In summary, while exterior shading systems and dynamic glazing could reduce disturbing glare
while working individually, the combination systems could reduce all the perceptible glare.
However, the EC glazing with kinetic horizontal blinds or static overhang failed to solve the
intolerable glare issue at 5 pm when the sky luminance was extremely high due to the sunset, the
system combining EC glazing and kinetic miniature blinds was able to reduce it to imperceptible
glare. Test condition #7 had the best performance in reducing glare, while test conditions #6 and
#8 ranked right after it.

122
All in all, test condition #7 had the best performance in improving glare issues. The glare was
imperceptible all day long. Test conditions #4, #5, #6, and #8 all had the second-best glare
performance according to 11 hours of imperceptible glare. They all had one intolerable glare
situation at 5 pm.
4.2.3 Peak Cooling Load
This section contained data on peak cooling loads. To measure the glazing and shading systems’
performance in reducing energy consumption, peak cooling loads were the most relevant factor.
It did not take heating trade-offs and other factors into account. The next section would measure
that. The table shows the peak cooling loads of all eight test conditions (Table 4-38).
Table 4-38 Peak Cooling Loads of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Peak Cooling Load (kW) 25.34 16.93 14.04 17.94 14.39 12.99 12.03 12.02
4.2.3.1 Test Condition #1 (D)
Test condition #1 had the highest value of the peak cooling load (FIG. 4-72). The cooling load
on the summer design day peaked at 3 pm, which was 25.34 kilowatts (kW).

Figure 4-72 Peak Cooling Load of Test Condition #1

123
4.2.3.2 Test Condition #2 (D+H)
Test condition #2 had a much lower value of peak cooling load, which was 16.93 kW (FIG. 4-
73). Different from the curve trend of the shoebox model, there was no obvious concave trend in
the afternoon. The cooling load peaked at 3 pm. The reason was that windows on the south and
west façades received a different amount of solar irradiance, so the kinetic blinds were not
deployed at the same time. Blinds on the south façade were deployed around noon, and the west
blinds were deployed when the sunset. They did not work together to reduce the solar heat
drastically at the same time.

Figure 4-73 Peak Cooling Load of Test Condition #2
4.2.3.3 Test Condition #3 (D+M)
Test condition #3 had a lower peak cooling load at 2 pm, which was 14.04 kW (FIG. 4-74). The
cooling loads of the whole day of the summer design day reduced a lot.

124

Figure 4-74 Peak Cooling Load of Test Condition #3
4.2.3.4 Test Condition #4 (D+O)
Test condition #4 had its cooling load peaked at 4 pm (FIG. 4-75). Since the overhang was not
kinetic, there was an obvious spike around 4 pm. The peak cooling load was 17.94 kW. The
value was higher than test conditions #2 and #3 but still improved from test condition #1.

Figure 4-75 Peak Cooling Load of Test Condition #4
4.2.3.5 Test Condition #5 (E)
Test condition #5 had its peak cooling load at 10 am (FIG. 4-76). The EC glazing shielded more
solar heat in the afternoon than in the morning. Therefore, there was a decrease in the trend line
from the peak at 10 am. The peak cooling load was 14.39 kW, which was a little higher than test
condition #3.

125

Figure 4-76 Peak Cooling Load of Test Condition #5
4.2.3.6 Test Condition #6 (E+H)
Test condition #6 has its peak cooling load was 12.99 kW at 5 pm (FIG. 4-77). It took 5.1 hours
to simulate because of the complexity of the model. The trend line was complicated. It had a
sunken point around 2 pm, and then the cooling load increased again at 5 pm.

Figure 4-77 Peak Cooling Load of Test Condition #6
4.2.3.7 Test Condition #7 (E+M)
Test condition #7 had a similar trend line to test condition #6 (FIG. 4-78). However, it was
overall decreased from the trend line of test condition #6. The peak cooling load was 12.03 kW
at 5 pm. It was the second-lowest value.

126

Figure 4-78 Peak Cooling Load of Test Condition #7
4.2.3.8 Test Condition #8 (E+O)
Test condition #8 had a similar trend line to test condition #4 (FIG. 4-79). However, the peak
point was a lot lower than test condition #4. The peak point was at 4 pm, and the peak cooling
load was 12.02 kW, which was the lowest peak cooling load.

Figure 4-79 Peak Cooling Load of Test Condition #8
In summation, all test conditions had their peak points at different times of the day. The
trendlines of the cooling loads on the summer design day varied a lot. Test condition #8 had the
lowest peak cooling load.

127
4.2.4 EUI
Different from peak cooling loads, EUI was correlated to annual cooling consumption, thermal
comfort, and daylight availability. The EUI results would be used to compare to the
benchmarking EUI to see whether the test conditions had met the performance goal of
Architecture 2030 that the energy consumption was 80% less than the local buildings’ EUI. The
table shows the breakdown of EUI of each test condition (Table 4-39).
Table 4-39 EUI Breakdowns of All Test Conditions
Test Conditions /
EUI (kWh/m
2
)
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Heating 0.1 0.37 0.75 0.22 0 0.03 0.07 0.04
Cooling 48.48 26.72 27.23 23.48 19.59 16.73 16.18 16.70
Fixed Interior Lighting 28.23 28.23 28.23 28.23 28.23 28.23 28.23 28.23
Interior Equipment 34.29 34.29 34.29 34.29 34.29 34.29 34.29 34.29
Original EUI 111.1 89.61 90.5 86.22 82.1 79.27 78.77 79.26
Lighting Deducted 17.75 17.26 11.79 13.07 7.69 6.59 6.24 2.90
New Interior Lighting 10.48 10.97 16.44 15.16 20.54 21.64 21.99 25.33
New EUI 93.35 72.35 78.71 73.15 74.42 72.69 72.53 76.36
4.2.4.1 Test Condition #1 (D)
Test condition #1 had the highest value of annual cooling consumption and EUI (Table 4-40).
The annual cooling consumption was 48.48 kWh/m
2
, and the EUI was 111.10 kWh/m
2
before
the extra artificial lighting was deducted from the flat interior lighting energy consumption. The
interior lighting deducted from the original EUI was 17.75 kWh/m
2
. The new EUI was 93.35
kWh/m
2
.
Table 4-40 EUI Breakdowns of Test Condition #1
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.1 48.48 28.23 34.29 111.10 17.75 93.35

128
4.2.4.2 Test Condition #2 (D+H)
Test condition #2 had a much lower value of annual cooling consumption and EUI, which were
26.72 kWh/m
2
and 89.61 kWh/m
2
respectively (Table 4-41). The interior lighting could be
reduced from the original EUI was 17.26 kWh/m
2
. The new EUI was 72.35 kWh/m
2
, which was
the lowest among all test conditions.
Table 4-41 EUI Breakdowns of Test Condition #2
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.37 26.72 28.23 34.29 89.61 17.26 72.35
4.2.4.3 Test Condition #3 (D+M)
Test condition #3’s annual cooling consumption and EUI were slightly higher than test condition
#2, which were 27.33 kWh/m
2
, and 90.50 kWh/m
2
(Table 4-42). The reason for higher EUI was
that test condition #3 had a higher annual heating consumption. However, the higher value of
annual cooling consumption might be the variation of simulations. The two values were almost
the same, so the abilities to reduce solar heat of the two test conditions were almost the same.
The artificial lighting reduced from the original EUI was 11.79 kWh/m
2
. This amount was lower
than the previous one because less sufficient daylighting was available. Hence, the new EUI was
higher than the previous test condition, which was 78.71 kWh/m
2
.
Table 4-42 EUI Breakdowns of Test Condition #3
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.75 27.23 28.23 34.29 90.50 11.79 78.71

129
4.2.4.4 Test Condition #4 (D+O)
For test condition #4, the cooling consumption was 23.48 kWh/m
2
and the EUI was 86.22
kWh/m
2
(Table 4-43). Those values were higher than test conditions #2 and #3 but still improved
from test condition #1. The amount of electric lighting that could be deducted from the original
EUI was 13.07 kWh/m
2
. It was higher than test condition #3 but lower than test condition #2.
The new EUI was 73.15 kWh/m
2
, which was a little higher than test condition #3.
Table 4-43 EUI Breakdowns of Test Condition #4
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.22 23.48 28.23 34.29 86.22 13.07 73.15
4.2.4.5 Test Condition #5 (E)
Test condition #5 had its annual cooling consumption reduced a lot to 19.59 kWh/m
2
from test
condition #4 (Table 4-44). The EUI was 82.10 kWh/m
2
, which was lower than all previous test
conditions. The interior lighting that could be removed was 7.69 kWh/m
2
. The new EUI was
74.42 kWh/m
2
.
Table 4-44 EUI Breakdowns of Test Condition #5
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0 19.59 28.23 34.29 82.10 7.69 74.42
4.2.4.6 Test Condition #6 (E+H)
The annual cooling consumption of test condition #6 was 16.73 kWh/m
2
and the EUI was 79.29
kWh/m
2
(Table 4-45). The artificial lighting could be reduced was 6.59 kWh/m
2
, and the new
EUI was 72.69 kWh/m
2
.

130
Table 4-45 EUI Breakdowns of Test Condition #6
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.03 16.73 28.23 34.29 79.27 6.59 72.69
4.2.4.7 Test Condition #7 (E+M)
The annual cooling consumption of test condition #7 was 16.18 kWh/m
2
, and the EUI was 78.77
kWh/m
2
(Table 4-46). The artificial lighting could be reduced was 6.24 kWh/m
2
. The new EUI
was 72.53 kWh/m
2
, which was much improved as well. It was the second-lowest EUI among all
test conditions.
Table 4-46 EUI Breakdowns of Test Condition #7
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.07 16.18 28.23 34.29 78.77 6.24 72.53
4.2.4.8 Test Condition #8 (E+O)
The EUI of test condition #8 was 79.26 kWh/m
2
, and the electric lighting deducted from the EUI
was 2.90 kWh/m
2
(Table 4-47). The new EUI was 76.36 kWh/m
2
.
Table 4-47 EUI Breakdowns of Test Condition #8
Heating
(kWh/m
2
)
Cooling
(kWh/m
2
)
Interior
Lighting
(kWh/m
2
)
Interior
Equipment
(kWh/m
2
)
Original
EUI
(kWh/m
2
)
Lighting
Deducted
(kWh/m
2
)
New EUI
(kWh/m
2
)
0.04 16.70 28.23 34.29 79.26 2.90 76.36
To sum up, the energy performance varied for different categories. While test condition #7 (EC
glazing + kinetic 45° miniature blinds) had the lowest peak cooling load, annual cooling
consumption, and original EUI value. Test condition #2 (clear double-pane glazing) had the
lowest new EUI value with daylight harvesting. Test condition #7 had the second-lowest new

131
EUI value when the interior lighting with sufficient daylight was deducted. Therefore, test
condition #2 had the best result of EUI among all test conditions.
4.2.5 Adaptive Thermal Comfort
The non-air-conditioned and non-naturally ventilated situations of test conditions #1 and #4 were
recorded as the passive strategies to reduce solar heat. However, the results showed it could be
dangerous and deadly to human beings. Therefore, the HVAC systems were necessary for the
Entrada project which had so many window areas to absorb solar heat. Other test conditions were
not discussed because they were considered active shading and glazing strategies. The image
shows the neutral temperatures over the year (FIG. 4-80).

Figure 4-80 Neutral Temperatures of the Year
4.2.5.1 Test Condition #1 (D)
If there was no HVAC system, the whole time was overheated (FIG. 4-81). The highest indoor
temperature was 74.55 °C, which was 51.70 °C higher than the neutral temperature (Table 4-48).
It was deadly to human beings. The image shows deviation from neutral temperatures (FIG. 4-
82). The large deviations in temperatures mostly happened in the afternoon of autumn and winter
which might cause by the low-angle sun. However, the simulation results were only estimates,
not the actual temperatures without the air-conditioning. The Entrada building may have
different glazing systems or other approaches to lower the interior temperatures.

132

Figure 4-81 Adaptive Thermal Comfort of Test Condition #1

Table 4-48 Thermal Information of Test Condition #1
%
Hot
% Neutral
(Comfortable)
%
Cold
Lowest
Temperature
°C
Highest
Temperature
°C
Lowest
Deviation °C
Highest
Deviation °C
100 0 0 25.07 74.55 3.02 51.70

Figure 4-82 Deviation from Neutral Temperatures of Test Condition #1
4.2.5.2 Test Condition #4 (D+O)
When there were static overhangs, still most of the year was overheated (FIG. 4-83). The highest
indoor temperature was 56.04 °C, which was 33.80 °C higher than the neutral temperature (Table
4-49). Although improved a lot from test condition #1, it was still deadly to human beings. High
deviation temperatures still appeared mostly in autumn and winter (FIG. 4-84).

Figure 4-83 Adaptive Thermal Comfort of Test Condition #1

133

Table 4-49 Thermal Information of Test Condition #4
%
Hot
% Neutral
(Comfortable)
%
Cold
Lowest
Temperature
°C
Highest
Temperature
°C
Lowest
Deviation °C
Highest
Deviation °C
99.97 0.03 0 22.40 56.04 0.34 33.80

Figure 4-84 Deviation from Neutral Temperatures of Test Condition #4
4.2.6 PMV Thermal Comfort
The indicator of thermal comfort measured the PMV thermal comfort when the room was air-
conditioned, and windows were not operable for ventilation. The Entrada building does not have
operable windows, and it was all relied on the HVAC system. The default setpoints of the large
office building type were 21 °C for heating and 24 °C for cooling from 6 am to 9 pm. For the rest
of the unoccupied hours, the heating setpoint was 15.6 °C and the cooling setpoint was 26.7 °C.
When the room was not air-conditioned, the highest temperature went up to 70 °C based on the
simulation results, which were fatal to human beings. Therefore, there was still a large amount of
overheated time when the HVAC system with default setpoints was used. This section records
data and simulation results of PMV thermal comfort conditions for different shading and glazing
options. The percentage of overheated time was the indicator for the PMV thermal performance
of the eight test conditions. The cold time was mostly caused by the low heating setpoint during
the unoccupied hours, so it would be excluded from the scoring system. Also, the cold
percentage of time reduced the percentage of the comfortable time, so only the percentage of
overheated time was useful for the evaluation. The table below shows the PMV thermal comfort

134
of all test conditions (Table 4-50).
Table 4-50 PMV Thermal Conditions of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% Hot 36.10 21.19 3.85 15.99 4.01 0.41 0 0.30
% Comfortable 59.01 69.57 79.42 74.87 93.95 96.26 92.77 95.94
% Cold 4.89 9.25 16.74 9.13 2.04 3.33 7.22 3.77
4.2.6.1 Test Condition #1 (D)
Test condition #1 usually had the worst thermal condition (FIG. 4-85). When the HVAC system
was on, 59.01% of the time was comfortable, and 36.10% was overheated time (Table 4-51).
While there were lots of hot conditions in summer, there were also a lot of overheated situations
in winter brought by the low-angle sun.

Figure 4-85 Thermal Condition of Test Condition #1

Table 4-51 Thermal Condition of Test Condition #1
% Too Hot % Comfortable % Too Cold
36.10 59.01 4.89
4.2.6.2 Test Condition #2 (D+H)
Test condition #2 had a more uncomfortable time in summer and autumn (FIG. 4-86). There was
21.19% overheated time and 69.57% comfortable time (Table 4-52). The situation improved
from the previous one.

135

Figure 4-86 Thermal Condition of Test Condition #2

Table 4-52 Thermal Condition of Test Condition #2
% Too Hot % Comfortable % Too Cold
21.19 69.57 9.25
4.2.6.3 Test Condition #3 (D+M)
Test condition #3 only had 3.85% overheated time (FIG. 4-87). 79.42% of the time was
comfortable. Although there was more percentage of comfortable time, there was more too cold
time as well, which was 16.74% (Table 4-53).

Figure 4-87 Thermal Condition of Test Condition #3

Table 4-53 Thermal Condition of Test Condition #3
% Too Hot % Comfortable % Too Cold
3.85 79.42 16.74
4.2.6.4 Test Condition #4 (D+O)
Test condition #4 had 15.99% overheated time and 74.87% comfortable time (Table 4-54). The
percentage of overheated time in the summer was reduced from test conditions #1 and #2 (FIG.
4-88). The comfortable time was a little less than test condition #3.

136

Figure 4-88 Thermal Condition of Test Condition #4

Table 4-54 Thermal Condition of Test Condition #4
% Too Hot % Comfortable % Too Cold
15.99 74.87 9.13
4.2.6.5 Test Condition #5 (E)
Test condition #5 had 4.01% overheated time and 93.95% comfortable time (Table 4-55). There
was 2.04% too cold time over the year but was less than the previous test condition. It had a lot
more comfortable time than the previous test conditions (FIG. 4-89).

Figure 4-89 Thermal Condition of Test Condition #5

Table 4-55 Thermal Condition of Test Condition #5
% Too Hot % Comfortable % Too Cold
4.01 93.95 2.04
4.2.6.6 Test Condition #6 (E+H)
Test condition #6 had most of the comfortable percentage among all test conditions (FIG. 4-90),
which was 96.26% (Table 4-56). The overheated time was little, which was 0.41% only, which
happened mostly in September. The too cold time was 3.33%.

137

Figure 4-90 Thermal Condition of Test Condition #6

Table 4-56 Thermal Condition of Test Condition #6
% Too Hot % Comfortable % Too Cold
0.41 96.26 3.33
4.2.6.7 Test Condition #7 (E+M)
Test condition #7 had no overheated time, and 92.77% of comfortable situations (Table 4-57).
There was more percentage of cold time, which was 7.22% because kinetic miniature blinds
caused more cold time. There was no uncomfortable time at all from June to September (FIG. 4-
91).

Figure 4-91 Thermal Condition of Test Condition #7

Table 4-57 Thermal Condition of Test Condition #7
% Too Hot % Comfortable % Too Cold
0 92.77 7.22
4.2.6.8 Test Condition #8 (E+O)
When the room had the HVAC system, the overheated time of test condition #8 was 0.30%, and
the comfortable time was 95.94%, which was the highest among all test conditions (Table 4-58).

138
There was no uncomfortable time from May to June (FIG. 4-92).

Figure 4-92 Thermal Condition of Test Condition #8

Table 4-58 Thermal Condition of Test Condition #8
% Too Hot % Comfortable % Too Cold
0.30 95.94 3.77
In summation, when the HVAC system was used, test condition #7 successfully reduced all
overheated conditions for a whole year, while the baseline value of overheated conditions
brought by test condition #1 was 36.10%.
4.2.7 PPD
This section records the PPD value for the worst overheated conditions and coldest conditions,
and the highest and lowest indoor temperatures of the year. The image shows the scale of PPD
was from 5% to 20% (FIG. 4-94). The table shows the temperatures and PPD values of all test
conditions (Table 4-59). Since the coldest conditions were mainly caused by the low heating
setpoint during the unoccupied hours, only the PPD values of the hottest time would be used to
evaluate the thermal performance in ameliorating uncomfortable levels.

139

Figure 4-93 Scale of PPD

Table 4-59 PPD and Indoor Temperatures of All Test Conditions
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Lowest °C 16.8 16.75 16.71 16.26 20.59 20.17 19.13 19.16
Highest °C 29.41 27.21 26.4 27.80 26.65 26.11 25.76 26.22
PPD % for the Hottest
Time
54.58 20.63 17.01 28.59 16.85 12.91 9.33 13.45
PPD % for the
Coldest Time
85.04 84.75 85.27 89.31 39.81 44.10 53.56 45.75
4.2.7.1 Test Condition #1 (D)
Test condition #1 had a very high PPD value. The highest indoor temperature was 29.41 °C
(Table 4-60), which was overheated. The temperature variation was 12.61 °C. The highest PPD
value for the overheated time of the year was 54.58%, which was much higher than the target of
20%. The most extreme situation happened in summer and autumn (FIG. 4-94).

Figure 4-94 PPD of Test Condition #1

140
Table 4-60 PPD and Temperatures of Test Condition #1
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.8 29.41 54.58 85.04
4.2.7.2 Test Condition #2 (D+H)
The highest indoor temperature was 27.21 °C (Table 4-61), which was lower than test condition
#1. The PPD value for the worst overheated conditions was 20.63%, which was only a little
higher than the 20% target, but still indicated overheating. The hot conditions happened in the
afternoon of summer and autumn (FIG. 4-95). The highest PPD decreased a lot from test
condition #1.

Figure 4-95 PPD of Test Condition #2

Table 4-61 PPD and Temperatures of Test Condition #2
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.75 27.21 20.63 84.75
4.2.7.3 Test Condition #3 (D+M)
The highest indoor temperature was 26.40 °C (Table 4-62), and the PPD for the hottest time was
17.01%. The worst overheated time happened in early June and early August (FIG. 4-96). The
overheated condition was below the 20% threshold, so it was acceptable.

141

Figure 4-96 PPD of Test Condition #3

Table 4-62 PPD and Temperatures of Test Condition #3
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.71 26.40 17.01 85.27
4.2.7.4 Test Condition #4 (D+O)
The highest indoor temperature was 27.80 °C (Table 4-63), increased from test condition #3. The
highest PPD value for the overheated time also increased, which was 28.59%. The value was
higher than the previous test condition and was higher than the 20% threshold as well (FIG. 4-
97). Therefore, the overheated level was uncomfortable for occupants even with the HVAC
system.

Figure 4-97 PPD of Test Condition #4

Table 4-63 PPD and Temperatures of Test Condition #4
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
16.26 27.80 28.59 89.31

142
4.2.7.5 Test Condition #5 (E)
Test condition #5 had the highest temperature 26.65 °C (Table 4-64). The highest PPD of
overheated conditions dropped to 16.85%, which was lower than 20%, and was improved from
the previous test conditions (FIG 4-98). This meant EC glazing was good in adjusting indoor
temperatures. The hottest conditions happened in the afternoon of September.

Figure 4-98 PPD of Test Condition #5

Table 4-64 PPD and Temperatures of Test Condition #5
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
20.59 26.65 16.85 39.81
4.2.7.6 Test Condition #6 (E+H)
The highest temperature was 26.11 °C (Table 4-65), which was a little lower than test condition
#5. The PPD value for the hottest time was 12.91%, which was the second-lowest among all test
conditions. The uncomfortable time in the summer was largely reduced (FIG. 4-99). The noon
and afternoon of early September had the worst overheated conditions.

Figure 4-99 PPD of Test Condition #6

143

Table 4-65 PPD and Temperatures of Test Condition #6
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
20.17 26.11 12.91 44.10
4.2.7.7 Test Condition #7 (E+M)
The highest indoor temperature was 25.76 °C (Table 4-66), which was the lowest among all test
conditions. The highest PPD for the overheated condition was 9.33%, which was lower than
10%, so it was comfortable for occupants (FIG. 4-100). The strategy of EC glazing with kinetic
miniature blinds successfully solved overheating issues. The hottest time happened in the
morning of early September.

Figure 4-100 PPD of Test Condition #7

Table 4-66 PPD and Temperatures of Test Condition #7
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
19.13 25.76 9.33 53.56
4.2.7.8 Test Condition #8 (E+O)
For test condition #8, the highest indoor temperature was 26.22 °C (Table 4-67), and the PPD
value for the worst overheated condition was 13.45%, which was a little higher than test
conditions #5 and #6 (FIG. 4-101). The worst overheated time happened in the afternoon of early
September.

144

Figure 4-101 PPD of Test Condition #8

Table 4-67 PPD and Temperatures of Test Condition #8
Lowest °C Highest °C PPD % for the
Hottest Time
PPD % for the
Coldest Time
19.61 26.22 13.45 45.75
All in all, test condition #1 had the highest indoor temperature, which was disturbing to
occupants even with the air-conditioning. Test condition #1 had the highest PPD value for the
hottest conditions. Test condition #7 had the lowest PPD value, so it was the most effective test
condition to reduce overheated conditions.
4.3 Conclusion
In conclusion, since the shoebox model and the Entrada project had different setups of apertures,
walls, and columns, the two of them had different performance results. The eight test conditions
had different ranks for the two situations. However, test conditions #6, #7, and #8, which were
the combined systems of dynamic glazing and exterior shading systems, had overall better
performance than other test conditions in most categories. See Chapter 5 for their scores and
detailed analyses.

145
CHAPTER FIVE ANALYSIS
The data of eight test conditions of both situations had been recorded and briefly summarized.
However, it was still hard to directly see which of the test condition had the best comprehensive
performance among all six categories and whether they had met the performance goals required
by different standards like ASHRAE 90.1 and 55, WELL Building Standard, and challenges in
Architecture 2030. A scoring system was developed for the test conditions for both the shoebox
model and the Entrada project. Each of the categories would have one test condition that had the
highest score, meaning there would be one shading or glazing strategy best for daylighting, glare,
reducing peak cooling loads, EUI, or maintaining thermal comfort respectively. There would
also be one test condition that had the best comprehensive performance among all categories for
both Experiments #1 and #2, meaning that the test condition had an overall good performance if
considering all factors including visual, energy, and thermal performance may conflict with each
other.
Test conditions had ranks according to each of the indicators, and then there would be a score for
each of the ranks. See the table for sample rubrics (Table 5-1). For instance, the first ranking
earned 8 points, and the last ranking earned 1 point. Take Daylight Glare Probability (DGP) as
an example, the smaller the DGP value, the better the performance of shielding glare. In this
example, test condition #1 had the least value of DGP, so it had the first ranking, thus earning 8
points. It also indicated test condition #1 had the best glare performance. When two test
conditions had the same performance, they had the same ranking and earned the same scores.
Take test conditions #5 and #6 as examples, they both had the fifth ranking and earned 4 points,
but the subsequent test condition #7 had the seventh ranking and earned only 2 points.

146
Table 5-1 Score and Rank Rules
Test
Conditions
#1 #2 #3 #4 #5 #6 #7 #8
DGP 10% 20% 30% 40% 50% 50% 70% 80%
DGP Rank 1 2 3 4 5 5 7 8
DGP Score 8 7 6 5 4 4 2 1
The scoring system had three categories of energy, visual, and thermal performance that had
equal weights. Each of the categories had two indicators. The visual comfort category had
average daylight performance and DGP. The energy category had peak cooling loads and EUI.
The thermal category had Predicted Mean Vote (PMV) thermal comfort and Predicted
Percentage of Dissatisfied (PPD) that assessed extreme discomfort situations. The sample table
shows the overall scoring and ranking rubrics (Table 5-2). In this example, test condition #1 had
the highest score, so it had the best performance if considering all categories.
Table 5-2 The Overall Scoring and Ranking Sample
Test Conditions #1 #2 #3 #4 #5 #6 #7 #8
Visual Score 1 8 7 6 5 4 3 2 1
Visual Score 2 8 7 6 5 4 3 2 1
Energy Score 1 8 7 6 5 4 3 2 1
Energy Score 2 8 7 6 5 4 3 2 1
Thermal Score 1 8 7 6 5 4 3 2 1
Thermal Score 2 8 7 6 5 4 3 2 1
Overall Score 48 42 36 30 24 18 12 6
Overall Rank 1 2 3 4 5 6 7 8
The extra evaluation included whether meeting the performance goals of 80% reduction of EUI,
Spatial Daylight Autonomy (sDA), DGP, indoor temperatures, and view qualities. However, they
would not part of the scoring system.

147
5.1 Experiment #1 - The Shoebox Model
The performance of the test conditions with combinatorial systems (test conditions #6, #7, and
#8) was not as effective as expected because there was only a small window on the west façade.
This meant the shoebox model did not have too many impacts from the sunlight and solar heat
through the window. Because of that, the Entrada model with two whole facades of glass had a
different test condition earning the highest score.
5.1.1 Daylighting
The percentage of areas with illuminance larger than 300 lux of each hour was listed in the table
(Table 5-3). The line chart shows the changes of the day (FIG. 5-1). The values in the morning
were similar. The values at 5 pm were the highest for all test conditions. The other values in the
afternoon varied.

Figure 5-1 Line Chart of Hourly Sufficient Daylighting
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
7:00 AM 8:00 AM 9:00 AM 10:00
AM
11:00
AM
12:00
PM
1:00 PM 2:00 PM 3:00 PM 4:00 PM 5:00 PM 6:00 PM
Momentary % ≥ 300 lux on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

148

Table 5-3 Sufficient Daylighting of Each Hour
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6 E+H #7
E+M
#8 E+O
7:00 AM 11.46% 11.46% 11.46% 6.77% 11.46% 10.94% 11.46% 4.17%
8:00 AM 14.58% 14.58% 14.58% 7.81% 13.02% 11.98% 12.50% 7.29%
9:00 AM 15.63% 15.63% 15.63% 8.33% 13.02% 13.54% 14.06% 7.29%
10:00 AM 16.67% 15.63% 16.15% 8.85% 14.06% 14.06% 14.06% 7.29%
11:00 AM 17.71% 17.71% 17.71% 9.38% 16.15% 16.15% 16.15% 8.33%
12:00 PM 19.27% 19.27% 19.27% 11.46% 4.69% 5.21% 4.69% 0.00%
1:00 PM 23.96% 23.44% 24.48% 14.58% 4.69% 4.69% 4.69% 0.52%
2:00 PM 27.60% 29.17% 28.13% 19.27% 9.38% 9.38% 9.38% 0.52%
3:00 PM 34.38% 28.65% 0.00% 23.96% 18.75% 4.69% 0.00% 9.38%
4:00 PM 40.10% 34.38% 0.00% 28.65% 27.60% 22.92% 0.00% 18.23%
5:00 PM 57.29% 57.81% 57.29% 55.21% 15.10% 15.63% 15.63% 4.17%
6:00 PM 13.02% 12.50% 14.06% 8.33% 12.50% 11.46% 11.98% 6.77%
The average values were recorded in the second table (Table 5-4). The highest value earned the
highest score. The bar chart shows the average percentage of areas with illuminance larger than
300 lux on the fall equinox day (FIG. 5-2). The percentage value decreased from test condition
#1 to test condition #8. Test condition #1 had the highest score, and test condition #8 had the
lowest score. Hence, test condition #1 had the best performance of daylighting.

Figure 5-2 Bar Chart of Average Sufficient Daylighting
24.31%
23.35%
18.23%
16.88%
13.37%
11.72%
9.55%
6.16%
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
% ≥ 300 lux on Fall Equinox

149

Table 5-4 Daylighting Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% ≥ 300 lux on
9/21 (Indicator)
24.31% 23.35% 18.23% 16.88% 13.37% 11.72% 9.55% 6.16%
sDA300,50%
on 9/21
33.77% 32.12% 27.17% 25.52% 17.45% 15.71% 14.06% 9.64%
Daylight Rank 1 2 3 4 5 6 7 8
Daylight Score 8 7 6 5 4 3 2 1
5.1.2 Glare
The table records the DGP of each occupied hour on the fall equinox day (Table 5-5). The line
chart shows the trend of the data (FIG. 5-4). The 3D line chart shows overlapped data (FIG. 5-3).
In the morning, DGP was low. The trendlines are flatter and smoother. In the afternoon, the
trendlines of test conditions #1 to #4 peaked. Test conditions #5 to #8 had low points at 3 and 4
pm. At 5 pm, their trendlines peaked as well, but the highest points were much lower.

150

Figure 5-3 Line Chart of Hourly DGP
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
7:00 AM 8:00 AM 9:00 AM 10:00
AM
11:00
AM
12:00 PM 1:00 PM 2:00 PM 3:00 PM 4:00 PM 5:00 PM 6:00 PM
Daylight Glare Probability (DGP) on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

151

Figure 5-4 3D Line Chart of Hourly DGP

Table 5-5 Hourly DGP
Test
Conditions
#1 D #2 D+H #3
D+M
#4 D+O #5 E #6 E+H #7
E+M
#8 E+O
7:00 AM 25.75% 25.75% 25.75% 24.85% 25.09% 25.09% 25.09% 24.25%
8:00 AM 29.66% 29.66% 29.66% 28.91% 28.59% 28.59% 28.59% 27.88%
9:00 AM 34.24% 34.24% 34.24% 33.77% 31.10% 31.10% 31.10% 30.55%
10:00 AM 36.70% 36.70% 36.70% 36.27% 35.14% 35.14% 35.14% 34.72%
11:00 AM 38.44% 38.44% 38.44% 37.82% 36.86% 36.86% 36.86% 36.29%
12:00 PM 40.34% 40.34% 40.34% 39.11% 22.21% 22.21% 22.21% 21.79%
1:00 PM 45.38% 45.37% 45.38% 41.12% 17.55% 17.55% 17.55% 16.60%
2:00 PM 48.98% 48.98% 48.98% 44.55% 3.91% 3.91% 3.91% 2.82%
3:00 PM 100.00
%
100.00
%
21.39% 45.76% 49.37% 49.18% 0.48% 3.62%
4:00 PM 100.00
%
100.00
%
20.79% 100.00
%
55.25% 45.26% 0.45% 55.28%
5:00 PM 100.00
%
100.00
%
100.00
%
100.00
%
62.04% 62.04% 62.04% 62.00%
6:00 PM 28.74% 28.74% 28.74% 28.20% 28.02% 28.02% 28.02% 27.54%
1 Double-pane Glazing
3 Kinetic Miniature Blinds
5 EC Glazing
7 EC + Kinetic Miniature Blinds
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Daylight Glare Probability (DGP) on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

152
The second glare table records hours of glare status (Table 5-6). Test conditions #1 and #2 had
the most intolerable glare hours and least imperceptible glare hours. Test condition #7 had the
most imperceptible glare hours and least intolerable glare hours. The bar chart shows the
performance of glare hours visually (FIG. 5-5). However, glare hours were hard to rank and keep
scores. The average DGP value was calculated to indicate the overall glare score and rank, but
the values of DGP themselves did not mean anything, since DGP only indicates momentary
performance. The test condition with the lowest average DGP value had the best performance of
glare. Therefore, test condition #7 was the most effective solution to glare for the shoebox
model.

Figure 5-5 Bar Chart of Glare Situations

4 4
6
4
7 7
9 9
2 2
2
3
2 2
2
1
1 1
1
2
5 5
3 3 3 3
1
2
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
Daylight Glare Probability
Imperceptible Perceptible Disturbing Intolerable

153
Table 5-6 Glare Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Imperceptible Glare (hrs.) 4 4 6 4 7 7 9 9
Perceptible Glare (hrs.) 2 2 2 3 2 2 2 1
Disturbing Glare (hrs.) 1 1 1 2 0 0 0 0
Intolerable Glare (hrs.) 5 5 3 3 3 3 1 2
Avg. DGP 52.35
%
52.35
%
39.20
%
46.70
%
32.93
%
32.08
%
24.28
%
28.61
%
Glare Rank 7 7 5 6 4 3 1 2
Glare Score 2 2 4 3 5 6 8 7
5.1.3 Peak Cooling Loads
The bar chart shows the peak cooling loads of each test condition (FIG. 5-6). Test condition #1
had the highest peak cooling load and test condition #7 had the lowest peak cooling load. The
line chart shows the difference in the cooling loads on the summer design day August 21
st
(FIG.
5-7). Test condition #1 peaked at 4 pm and test condition #7 peaked at 2 pm. For reducing the
peak cooling load, test condition #7 had the best performance as shown in the table (Table 5-7).

Figure 5-6 Bar Chart of Peak Cooling Loads Figure 5-7 Highest and Lowest Peak Cooling Loads

2.55
2.06
1.91
2.16
1.79
1.55
1.44
1.71
0
0.5
1
1.5
2
2.5
3
Peak Cooling Loads (kilowatts)
16, 2.55
14, 1.44
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 2 4 6 8 10121416182022
Peak Cooling Loads
(kilowatts)
#1 #7

154
Table 5-7 Peak Cooling Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Peak Cooling Load (kW) 2.55 2.06 1.91 2.16 1.79 1.55 1.44 1.71
Peak Cooling Rank 8 6 5 7 4 2 1 3
Peak Cooling Score 1 3 4 2 5 7 8 6
5.1.4 EUI
The bar chart shows the breakdowns of EUI (FIG. 5-8). The cooling EUI affected most of the
total EUI. The electric lighting also did. The differences in heating were little. The equipment
was assumed to be the same. For all factors considered, the EUI value presented the energy
efficiency of all. Test condition #7 had the lowest EUI value, which meant it had the best
performance and the highest score in EUI as shown in the table (Table 5-8).

Figure 5-8 Bar Chart of EUI Breakdowns

34.32 34.32 34.32 34.32 34.32 34.32 34.32 34.32
21.38 21.65
23.09 23.47
24.46 24.93 25.54
26.50
16.44 15.86
14.58
11.34 9.72
9.43
7.29
9.03
1.74 1.79 1.85
2.32
1.91
2.43
3.30
2.20
30.00
35.00
40.00
45.00
50.00
55.00
60.00
65.00
70.00
75.00
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
New Energy Use Intensity (EUI) (kWh/m
2
)
Interior Equipment Interior Lighting Cooling Heating

155
Table 5-8 EUI Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Heating 1.74 1.79 1.85 2.32 1.91 2.43 3.30 2.20
Cooling 16.44 15.86 14.58 11.34 9.72 9.43 7.29 9.03
Interior Lighting 21.38 21.65 23.09 23.47 24.46 24.93 25.54 26.50
Interior
Equipment
34.32 34.32 34.32 34.32 34.32 34.32 34.32 34.32
New EUI
(kWh/m
2
)
73.86 73.61 73.84 71.45 70.41 71.05 70.39 72.05
EUI Rank 8 6 7 4 2 3 1 5
EUI Score 1 3 2 5 7 6 8 4
5.1.5 PMV Thermal Comfort
The bar chart shows the data of PMV thermal comfort (FIG. 5-9). The table shows the PMV
thermal comfort % and scores (Table 5-9). Test condition #1 had the highest percentage of
overheated conditions. All other test conditions did not have any overheated conditions. In this
case, test conditions #2 to #8 had tied PMV thermal performance, so they all earned eight points.
Test condition #1 had the lowest rank and earned only one point.

Figure 5-9 Bar Chart of PMV Thermal Comfort

72% 72% 71% 68% 67% 63% 56% 65%
27% 29%
29%
32% 33%
37%
44%
35%
0.64%
0% 0% 0% 0% 0% 0% 0%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
PMV Thermal Comfort
% Neutral % Cold % Hot

156
Table 5-9 PMV Thermal Comfort Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% Hot (Indicator) 0.64 0 0 0 0 0 0 0
% Cold 27.29 28.50 29.41 32.25 33.01 37.23 43.87 35.38
% Comfortable 72.07 71.50 70.59 67.75 66.99 62.77 56.13 64.62
PMV Thermal Rank 8 1 1 1 1 1 1 1
PMV Thermal Score 1 8 8 8 8 8 8 8
5.1.6 PPD
The second indicator of thermal comfort was the Predicted Percentage Dissatisfied (PPD) value
of the worst overheated conditions. If most of the indoor temperatures were comfortable, but the
rest of the time had extremely uncomfortable temperatures, the thermal comfort level decreased.
Therefore, PPD was an important indicator. The bar chart shows the extreme PPD value for the
hot time (FIG. 5-10). The table shows the value of the highest PPD and the scores (Table 5-10).
Test condition #7 had the best performance in reducing overheated conditions and maintaining
comfort levels.

Figure 5-10 Bar Chart of PPD

11.93%
10.00%
9.68% 9.77%
7.57%
6.94%
6.62%
7.11%
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
PPD % for the Hottest Time

157
Table 5-10 PPD Score
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6 E+H #7
E+M
#8 E+O
PPD %
(Overheated)
11.93% 10.00% 9.68% 9.77% 7.57% 6.94% 6.62% 7.11%
PPD Rank 8 7 5 6 4 2 1 3
PPD Score 1 2 4 3 5 7 8 6
5.1.7 Performance Goals
This section evaluated whether the test conditions had met certain standards, challenges, or
performance goals. However, this section was not part of the scoring system.
The first performance goal was Architecture 2030 Challenge. If the buildings could have an 80%
energy reduction compared to the regional average, the challenge was completed (Architecture
2030). Therefore, the challenge was to get the EUI 80% lower than the regional benchmarking
EUI of the same building type. Los Angeles was climate zone 3B, dry and warm. EUI of climate
zone 3B (CA) in National Renewable Energy Laboratory Evaluation of ASHRAE 90.1-2009 was
95 kWh/m
2
, which was a benchmark for commercial building instead of a regional average. This
research used CBECS 2012 data as the regional EUI of office building type in the dry and hot
climate. The value was 185 kWh/m
2
, so the performance goal of reducing 80% of energy use
from it would be 37 kWh/m
2
. The table below shows the EUI data of eight test conditions and
whether they achieved the goal (Table 5-11). For the shoebox model, none of the test conditions
achieved that performance goal.
Table 5-11 EUI Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
New EUI (kWh/m
2
) 73.86 73.61 73.84 71.45 70.41 71.05 70.39 72.05
≤ 37 kWh/m
2
?
✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕

158
The second goal was from WELL Building Standard v1 suggesting buildings have Spatial
Daylight Autonomy (sDA) of at least 55% (WELL Building Standard v1). The annual sDA data
of the dynamic shading or glazing system was not available for the Ladybug Tools to simulate,
so if the sDA on the fall equinox was larger than 55%, the performance goal was approximately
met. sDA was the 50% time of percentage of areas receiving sufficient daylighting. The table
shows the sDA values of eight test conditions on the fall equinox (Table 5-12). However, none
of the test conditions helped the shoebox model to achieve this goal because of the small ratio of
the window.
Table 5-12 sDA Performance Goal
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
sDA (300lux50%)
on 9/21
33.77% 32.12% 27.17% 25.52% 17.45% 15.71% 14.06% 9.64%
≥ 55%?
✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕
The third goal was to achieve 0% Spatial Glare Autonomy (sGA40%5%) developed by Nathaniel L
Jones from Arup described in Chapter 2 (Jones 2019). If no DGP was larger than 40% for more
than 5% of the occupied hours, the glare performance target was met. If there was one hour of
DGP larger than 40%, which meant disturbing and intolerable glare, then the goal was not met. If
all hours had DGP smaller than 40%, then the goal was met. The table (Table 5-13) shows that
no test condition completely excluded the disturbing or intolerable glare situations for the
shoebox model.

159
Table 5-13 sGA Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Disturbing Glare (hrs.) 1 1 1 2 0 0 0 0
Intolerable Glare (hrs.) 5 5 3 3 3 3 1 2
Hours of DGP ≥ 40% 6 6 4 5 3 3 1 2
< sGA40%5%?
✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕
The fourth performance goal was recommended by ASHRAE 55. The indoor temperatures were
recommended to fall within the range from 67 °F to 82 °F (ANSI/ASHRAE 2017). None of the
test conditions had annual indoor temperatures that completely fell into that range as shown in
the table (Table 5-14). The large office building had a heating setpoint of 15.6 °C during the
unoccupied hours. It decreased the lowest indoor temperatures of the year. However, this did not
mean the test conditions not meeting the goals had an uncomfortable thermal environment. It
only indicated that the heating system was still necessary for the indoor environment of the
shoebox model.
Table 5-14 Temperature Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Lowest °C 16.8 16.8 16.8 16.64 16.85 16.56 16.04 16.63
Highest °C 25.88 25.59 25.55 25.5 25.08 24.96 24.78 24.95
19.4 °C - 27.8 °C?
✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕
The last one was visual quality introduced by Won Hee Ko. If the test conditions could provide
clear and accessible views, the goal was met (Ko 2021). All test conditions had hours providing
clear and accessible views. The table below shows the dynamic glazing and shading stats (Table
5-15). Even test condition #7 had clear and accessible views in the morning and evening.
Therefore, all test conditions had provided at least hours of complete views of a day and
achieved the performance goal as shown in the table (Table 5-16).

160
Table 5-15 Dynamic States of the Shoebox Model

Table 5-16 View Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Clear View
✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Accessible View
✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
5.1.8 Final Scores and Ranks
The overall scores and ranks were shown in the table (Table 5-17). The one earning the highest
score in the shoebox model experiment was test condition #7, the strategy of using EC glazing
with kinetic miniature blinds with 45° tilt angles. It had the best overall performance among all
categories of energy, visual, and thermal comfort. Test condition #6 had the second-best
performance, which was EC glazing with kinetic horizontal blinds. The PMV thermal scores
were tied because they all had zero percentage overheated time. The tied scores situation was one
of the limitations of the scoring system, but the system did the assessment of the eight conditions
and figured out that one test condition had overall better performance than others. The result of
the performance scores was in accordance with the hypothesis that the combinatorial system had
overall better performance in all categories.

Time \ Facade 7 am 8 am 9 am 10 am 11 am 12 pm
West Clear Clear Clear Clear Clear Light
Time \ Facade 1 pm 2 pm 3 pm 4 pm 5 pm 6 pm
West Heavy Full Full &
Blinds
Full &
Blinds
Heavy Clear

161
Table 5-17 Overall Performance Score of the Shoebox Model
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Daylight Score 8 7 6 5 4 3 2 1
Glare Score 2 2 4 3 5 6 8 7
Peak Cooling Score 1 3 4 2 5 7 8 6
EUI Score 1 3 2 5 7 6 8 4
PMV Thermal Score 1 8 8 8 8 8 8 8
PPD Score 1 2 4 3 5 7 8 6
Overall Score 14 25 28 26 34 37 42 32
Overall Rank 8 7 5 6 3 2 1 4

162
5.2 Experiment #2 - The Entrada Project
The Entrada project had more window areas, so the room got a lot more sunlight and solar heat.
This would have a huge impact on all the performance of test conditions.
5.2.1 Daylighting
Because of large window areas on the west and south facades, test conditions #1 and #2 had a
high percentage of areas with sufficient daylighting. Other test conditions had changing
daylighting during the day. Overall, they all had a relatively large amount of daylight. The line
chart shows hourly data of sufficient daylighting percentage (FIG. 5-11). The table shows the
hourly value as well (Table 5-18).

Figure 5-11 Line Chart of Hourly Sufficient Daylighting

0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
7:00 AM 8:00 AM 9:00 AM 10:00
AM
11:00
AM
12:00
PM
1:00 PM 2:00 PM 3:00 PM 4:00 PM 5:00 PM 6:00 PM
Momentary % ≥ 300 lux on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

163
Table 5-18 Hourly Sufficient Daylighting
Test
Conditions
#1 D #2 D+H #3
D+M
#4 D+O #5 E #6 E+H #7
E+M
#8 E+O
7:00 AM 51.60% 52.19% 52.02% 35.44% 30.39% 30.39% 30.64% 17.85%
8:00 AM 60.52% 60.19% 61.20% 44.36% 19.95% 19.28% 19.53% 7.91%
9:00 AM 64.05% 58.42% 13.80% 47.05% 28.11% 16.41% 13.30% 9.01%
10:00 AM 64.90% 60.52% 15.15% 47.31% 28.11% 16.58% 14.23% 9.18%
11:00 AM 65.66% 60.61% 17.17% 47.64% 17.26% 7.74% 4.21% 0.17%
12:00 PM 65.82% 61.36% 20.03% 48.23% 19.61% 7.07% 7.07% 0.59%
1:00 PM 68.27% 68.10% 68.18% 49.24% 22.22% 22.22% 22.05% 3.20%
2:00 PM 70.71% 70.03% 70.20% 52.02% 22.90% 22.73% 22.73% 6.99%
3:00 PM 70.03% 71.80% 69.70% 53.79% 33.84% 33.84% 34.01% 12.63%
4:00 PM 69.70% 69.87% 68.86% 53.87% 57.07% 56.99% 56.73% 36.70%
5:00 PM 77.61% 74.75% 19.11% 65.99% 23.82% 23.65% 17.76% 10.77%
6:00 PM 25.84% 25.67% 25.84% 10.77% 23.65% 23.32% 22.98% 8.25%
For the average percentage of areas with sufficient daylight, test condition #1 had the highest
value, so it had the best daylighting performance. Test condition #8 had the lowest percentage of
areas with sufficient daylighting. The bar chart shows the average percentage of sufficient
daylighting received on the fall equinox day (FIG. 5-12). The table shows the value and the
scores (Table 5-19).

Figure 5-12 Bar Chart of Average Sufficient Daylighting
62.89%
61.13%
41.77%
46.31%
27.24%
23.35%
22.10%
10.27%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
% ≥ 300 lux on Fall Equinox

164

Table 5-19 Daylighting Score
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% ≥ 300 lux
on 9/21
(Indicator)
62.89% 61.13% 41.77% 46.31% 27.24% 23.35% 22.10% 10.27%
sDA300,50%
on 9/21
70.36% 69.32% 65.03% 53.86% 33.56% 31.82% 31.52% 16.02%
Daylight Rank 1 2 4 3 5 6 7 8
Daylight Score 8 7 5 6 4 3 2 1
5.2.2 Glare
The line chart shows the hourly DGP of September 21
st
(FIG. 5-13). The 3D line chart shows the
overlapped data (FIG. 5-14). Except for test conditions #3 and #7, other test conditions had a
peak point at 5 pm. The rest parts had relatively smooth trend lines. The table records all the
values of DGP (Table 5-20).

165

Figure 5-13 Line Chart of Hourly DGP
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
7:00 AM 8:00 AM 9:00 AM 10:00
AM
11:00
AM
12:00
PM
1:00 PM 2:00 PM 3:00 PM 4:00 PM 5:00 PM 6:00 PM
Daylight Glare Probability (DGP) on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

166

Figure 5-14 3D Line Chart of Hourly DGP

Table 5-20 Hourly DGP
Test
Conditions
#1 D #2 D+H #3
D+M
#4 D+O #5 E #6 E+H #7
E+M
#8 E+O
7:00 AM 28.20% 28.20% 28.20% 25.75% 20.20% 20.20% 20.20% 19.51%
8:00 AM 35.69% 35.69% 35.69% 26.88% 18.99% 18.99% 18.99% 18.69%
9:00 AM 38.77% 30.62% 24.59% 28.50% 19.55% 19.48% 19.32% 19.26%
10:00 AM 41.39% 31.63% 24.89% 29.15% 23.47% 23.49% 23.57% 23.56%
11:00 AM 40.52% 31.73% 25.30% 29.67% 15.73% 14.92% 12.67% 14.00%
12:00 PM 39.33% 31.39% 26.20% 30.34% 16.34% 16.04% 14.18% 14.77%
1:00 PM 38.18% 38.18% 38.18% 31.14% 17.34% 17.34% 17.34% 15.61%
2:00 PM 36.47% 36.47% 36.47% 32.12% 18.69% 18.69% 18.69% 17.90%
3:00 PM 36.35% 36.35% 36.35% 33.21% 21.21% 21.21% 21.21% 20.12%
4:00 PM 38.10% 38.10% 38.10% 34.62% 27.71% 27.71% 27.71% 26.26%
5:00 PM 100.00
%
100.00
%
22.25% 100.00
%
53.23% 52.38% 21.29% 53.44%
6:00 PM 26.76% 26.76% 26.76% 26.78% 26.20% 26.20% 26.20% 26.26%
1 Double-pane Glazing
3 Kinetic Miniature Blinds
5 EC Glazing
7 EC + Kinetic Miniature Blinds
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Daylight Glare Probability (DGP) on Fall Equinox
1 Double-pane Glazing 2 Kinetic Horizontal Blinds 3 Kinetic Miniature Blinds
4 Static Overhang 5 EC Glazing 6 EC + Kinetic Horizontal Blinds
7 EC + Kinetic Miniature Blinds 8 EC + Static Overhang

167
The bar chart shows hours of glare status (FIG. 5-15). Test condition #3 had no disturbing or
intolerable hours. Test condition #7 had all occupied hours of imperceptible hours. For the
average DGP, test condition #7 had the lowest value, so it had the best performance in reducing
glare as shown in the table (Table 5-21). The average DGP value itself did not mean anything. It
was used to evaluate the glare performance only. Test condition #8 had the second-best glare
performance.

Figure 5-15 Bar Chart of Glare Situations

Table 5-21 Glare Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Imperceptible Glare (hrs.) 2 6 7 11 11 11 12 11
Perceptible Glare (hrs.) 7 5 5 0 0 0 0 0
Disturbing Glare (hrs.) 2 0 0 0 0 0 0 0
Intolerable Glare (hrs.) 1 1 0 1 1 1 0 1
Avg. DGP (Indicator) 41.65
%
38.76
%
30.25
%
35.68
%
23.22
%
23.05
%
20.11
%
22.45
%
Glare Rank 8 7 5 6 4 3 1 2
Glare Score 1 2 4 3 5 6 8 7
2
6
7
11 11 11
12
11
7
5
5
2
1 1 1 1 1 1
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
Daylight Glare Probability
Imperceptible Perceptible Disturbing Intolerable

168
5.2.3 Peak Cooling Loads
The bar chart shows the peak cooling loads of all eight test conditions (FIG. 5-16). Test
condition #1 had the highest peak cooling load at 3 pm, and test condition #8 had the lowest peak
cooling load at 4 pm. The line chart shows the cooling loads on the summer design day of
August 21
st
of test conditions #1 and #8 to show the differences (FIG. 5-17). Test condition #8
had the best performance in reducing peak cooling loads as shown in the table (Table 5-22).

Figure 5-16 Bar Chart of Peak Cooling Loads Figure 5-17 Highest and Lowest Peak Cooling Loads

Table 5-22 Peak Cooling Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Peak Cooling Load
(kW)
25.34 16.93 14.04 17.94 14.39 12.99 12.03 12.02
Peak Cooling Rank 8 6 4 7 5 3 2 1
Peak Cooling Score 1 3 5 2 4 6 7 8
5.2.4 EUI
The bar chart shows the EUI breakdowns of eight test conditions (FIG. 5-18). All test conditions
had little heating EUI. Cooling EUI varied the most. The electric lighting EUI also varied. The
equipment EUI was assumed to be the same for all test conditions. Considering all the different
factors of heating, cooling, and lighting, test condition #2 had the lowest EUI value, so it had the
25.34
16.93
14.04
17.94
14.39
12.99
12.03 12.02
0
5
10
15
20
25
30
Peak Cooling Loads (kilowatts)
15, 25.34
16, 12.02
0.00
5.00
10.00
15.00
20.00
25.00
30.00
0 3 6 9 12 15 18 21
Peak Cooling Loads
(kilowatts)
#1 #8

169
best EUI performance. The table records all the EUI breakdowns and scores (Table 5-23).

Figure 5-18 Bar Chart of EUI Breakdowns

Table 5-23 EUI Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Heating 0.1 0.37 0.75 0.22 0 0.03 0.07 0.04
Cooling 48.48 26.72 27.23 23.48 19.59 16.73 16.18 16.7
Interior Lighting 10.48 10.97 16.44 15.16 20.54 21.64 21.99 25.33
Interior
Equipment
34.29 34.29 34.29 34.29 34.29 34.29 34.29 34.29
New EUI
(kWh/m
2
)
93.35 72.35 78.71 73.15 74.42 72.69 72.53 76.36
EUI Rank 8 1 7 4 5 3 2 6
EUI Score 1 8 2 5 4 6 7 3
5.2.5 PMV Thermal Comfort
The bar chart shows the thermal conditions of eight test conditions (FIG. 5-19) and the table
shows the thermal conditions and the thermal scores (Table 5-24). Different from the shoebox
model, the Entrada project had more overheated time than cold time because of large window
areas on the west and south facades. Test condition #1 had the highest percentage of time with
indoor temperatures falling into the overheated thermal range. Test condition #7 had the lowest
34.29 34.29 34.29 34.29 34.29 34.29 34.29 34.29
10.48 10.97
16.44 15.16
20.54 21.64 21.99
25.33
48.48
26.72
27.23
23.48
19.59 16.73 16.18
16.7
0.1
0.37
0.75
0.22 0 0.03 0.07 0.04
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
New Energy Use Intensity (EUI) (kWh/m
2
)
Interior Equipment Interior Lighting Cooling Heating

170
percentage of the overheated time. Therefore, test condition #7 had the best performance in
reducing the overheated time during the year.

Figure 5-19 Bar Chart of PMV Thermal Comfort

Table 5-24 Thermal Score
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
% Cold 4.89 9.25 16.74 9.13 2.04 3.33 7.22 3.77
% Hot (Indicator) 36.10 21.19 3.85 15.99 4.01 0.41 0 0.30
% Comfortable 59.01 69.57 79.42 74.87 93.95 96.26 92.77 95.94
PMV Thermal Rank 8 7 4 6 5 3 1 2
PMV Thermal Score 1 2 5 3 4 6 8 7
5.2.6 PPD
The bar chart shows the highest PPD value for the overheated time of eight test conditions (FIG.
5-20) and the table shows the highest PPD value and the PPD scores (Table 5-25). Test condition
#1 had the greatest value of Predicted Percentage Dissatisfied (PPD), which meant it did not
have a good performance in keeping the extreme overheated annual indoor temperature close to
the comfortable range. Test condition #7 had the smallest PPD value, so it had the best
performance in avoiding extremely uncomfortable thermal conditions.
59.01 69.57 79.42 74.87 93.95 96.26 92.77 95.94
36.1
21.19
3.85
15.99
4.01
0.41
0
0.3
4.89
9.25
16.74
9.13
2.04
3.33
7.22
3.77
50%
55%
60%
65%
70%
75%
80%
85%
90%
95%
100%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
PMV Thermal Comfort
% Neutral % Hot % Cold

171

Figure 5-20 Bar Chart of Highest PPD %

Table 5-25 PPD Score
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
PPD %
(Overheated)
54.58% 20.63% 17.01% 28.59% 16.85% 12.91% 9.33% 13.45%
PPD Rank 8 6 5 7 4 2 1 3
PPD Score 1 3 4 2 5 7 8 6
5.2.7 Performance Goals
The section evaluated whether any of the test conditions could meet any of the performance
goals in Experiment #2. The results would be largely different from Experiment #1.
The regional EUI benchmarking was 185 kWh/m
2
, and the value with a reduced 80% of EUI was
37 kWh/m
2
(Architecture 2030). If the EUI was smaller than or equal to 37 kWh/m
2
, then the test
conditions accomplished the Architecture 2030 Challenge. However, no test condition had met
the EUI goal as shown in the table (Table 5-26). This indicated that even though the building had
large glazing areas, improving the shading and glazing systems alone was not sufficient to
achieve the EUI goal. Perhaps installing the solar panels could compensate the EUI and lead to
54.58%
20.63%
17.01%
28.59%
16.85%
12.91%
9.33%
13.45%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
1 Double-
pane Glazing
2 Kinetic
Horizontal
Blinds
3 Kinetic
Miniature
Blinds
4 Static
Overhang
5 EC Glazing 6 EC + Kinetic
Horizontal
Blinds
7 EC + Kinetic
Miniature
Blinds
8 EC + Static
Overhang
PPD % for the Hottest Time

172
enough energy reduction and even achieve zero net energy (ZNE).
Table 5-26 EUI Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
New EUI
(kWh/m
2
)
93.35 72.35 78.71 73.15 74.42 72.69 72.53 76.36
≤ 37 kWh/m
2
?
✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕
If the sDA300lux50% on September 21
st
was larger than 55%, then the test conditions approximately
met this performance goal recommended by WELL Building Standard v1 (WELL Building
Standard v1). sDA was 50% time of the occupied hours received sufficient daylighting, which
was different from the average percentage of areas receiving sufficient daylighting. Test
conditions #1 to #4 had met the performance goal as shown in the table (Table 5-27). Test
conditions #5 to #8 did not meet the goal. However, this did not mean the test conditions actually
had sufficient sDA since it was only the data of a day.
Table 5-27 sDA Performance Goal
Test
Conditions
#1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
sDA 300lux50%
on 9/21
70.36% 69.32% 65.03% 53.86% 33.56% 31.82% 31.52% 16.02%
≥ 55%?
✓ ✓ ✓ ✓ ✕ ✕ ✕ ✕
If there were no disturbing or intolerable glare hours, which meant there were no DGP larger
than 40% for more than 5% occupied hours, then the test conditions met the spatial glare
autonomy (sGA40%5%) goal created by Jones from Arup (Jones 2019). Test conditions #3 and #7
did not have any disturbing or intolerable glare hours, so they approximately met the
performance goal. The table below shows the glare conditions (Table 5-28). However, it was still
possible that disturbing or intolerable glare appeared on some other day of the year since it was

173
only the data on the fall equinox.
Table 5-28 sGA Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Disturbing Glare (hrs.) 2 0 0 0 0 0 0 0
Intolerable Glare (hrs.) 1 1 0 1 1 1 0 1
Hours of DGP ≥ 40% 3 1 0 1 1 1 0 1
< sGA40%5%?
✕ ✕ ✓ ✕ ✕ ✕ ✓ ✕
If the indoor temperatures with the air-conditioning fell into the range from 19.4 °C to 27.8 °C,
then the test conditions met this performance goal of indoor temperatures recommended by
ASHRAE 55 (ANSI/ASHRAE 2017). Test conditions #5 and #6 had indoor temperatures that
fell within the recommended indoor temperature range. The table shows the indoor temperatures
of eight test conditions (Table 5-29). However, the large office building had a low heating
setpoint during the unoccupied hours that decreased the lower bound of the annual indoor
temperatures, so not meeting this performance goal did not mean the test conditions did not have
a comfortable thermal environment. This indicated that for some test conditions, the heating
system was still necessary. For test conditions #5 to #8, the heating system with a setpoint of
15.6 °C during the unoccupied hours was probably not used.
Table 5-29 Temperature Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Lowest °C 16.8 16.75 16.71 16.26 20.59 20.17 19.13 19.16
Highest °C 29.41 27.21 26.4 27.80 26.65 26.11 25.76 26.22
19.4 °C - 27.8 °C?
✕ ✕ ✕ ✕ ✓ ✓ ✕ ✕
The last one was whether the test conditions had clear and accessible views (Ko 2021). The table
below shows the dynamic shading and glazing states of the test conditions (Table 5-30). At least
one side of the glass façade had clear or light glazing even for test condition #7, so all test

174
conditions had clear and accessible views and met the goal developed by Won Hee Ko’s
assessment of view quality. The table shows that all test conditions had met that performance
goal (Table 5-31).
Table 5-30 Dynamic States of the Entrada Project
Time / Facade 7 am 8 am 9 am 10 am 11 am 12 pm
West Clear Clear Clear Clear Light Light
South Light Full Full &
Blinds
Full &
Blinds
Full &
Blinds
Full &
Blinds
Time / Facade 1 pm 2 pm 3 pm 4 pm 5 pm 6 pm
West Light Light Light Light Full &
Blinds
Clear
South Full Heavy Light Clear Clear Clear

Table 5-31 View Performance Goal
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Clear View
✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Accessible View
✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
5.2.8 Final Scores and Ranks
The table shows the overall scores and ranks of all eight test conditions (Table 5-32). The test
condition with the highest score for the Entrada project was test condition #7, and the second-
highest score was received by test condition #6, which were the same as the shoebox model. The
third-highest score was received by test condition #8, which was different from Experiment #1.
Test conditions #6, #7, and #8 were all combinatorial strategies that had EC glazing with kinetic
miniature blinds, kinetic horizontal blinds, and static overhang. This demonstrated that the
combinatorial system had better performance among energy, visual, and thermal categories than
other shading and glazing options, which was in accordance with the hypothesis. It was a
coincidence that test conditions #2 and #3 had tied scores, but they had different strengths in
different categories. For example, test condition #2 had better EUI and daylight scores, while test

175
condition #3 had better glare, cooling, and thermal scores.
Table 5-32 Overall Performance Score of the Entrada Project
Test Conditions #1 D #2
D+H
#3
D+M
#4
D+O
#5 E #6
E+H
#7
E+M
#8
E+O
Daylight Score 8 7 5 6 4 3 2 1
Glare Score 1 2 4 3 5 6 8 7
Cooling Score 1 3 5 2 4 6 7 8
EUI Score 1 8 2 5 4 6 7 3
PMV Thermal Score 1 2 5 3 4 6 8 7
PPD Score 1 3 4 2 5 7 8 6
Overall Score 13 25 25 21 26 34 40 32
Overall Rank 8 5 5 7 4 2 1 3
However, more of the test conditions had helped the building to meet the standards and
performance goals, including daylighting, glare, and thermal comfort. But the performance goal
of energy was not achieved. This meant the intervention on the glazing and shading systems was
not sufficient to reduce too much energy. More actions should be taken to meet the goal of
achieving energy performance.
In conclusion, for different building types, different values of WWR, and other different
situations, the best test conditions suitable for them were different. Additionally, all ranks and
scores of test conditions did not represent the performance of the entire shading or glazing
techniques. Test condition #5 did not represent all EC glazing and test condition #2 did not
represent all kinetic facades because they had specific forms and parameters. A single change of
SHGC or VLT of glazing, tilt angles of blinds, or controllers of the dynamic systems would lead
to completely different simulation results and performance in visual, energy, and thermal
comfort. This research mainly provided methods to compare the performance of visual, energy,
and thermal categories visually and digitally, and methods to put all categories into one scoring

176
system to evaluate the overall performance with the consideration of all the trade-offs. See
Chapter 6 for more discussions.

177
CHAPTER SIX DISCUSSION & FUTURE WORK
6.1 Discussion
The hypothesis stating that the combination of dynamic glazing and kinetic shade would have the
best performance among energy usage, daylight, glare, and thermal comfort was in accordance
with the experiment results for both the shoebox model and the Entrada project. The
combinatorial system of EC glazing and kinetic miniature blinds had the best overall
performance for both experiments. The strategy of EC glazing with kinetic horizontal blinds had
the second-best performance for both experiments. There were discrepancies in the scores and
other performance results. The discrepancy was mostly caused by the different sizes,
orientations, and positions of the glazing areas of the shoebox model and the Entrada project.
This chapter explained the reasons why different test conditions had different performance
among categories of visual, energy, and thermal comfort in two circumstances. The chapter also
covered future works that could be done that could refine the research.
Previous research from other scholars had covered the visual and thermal performance of EC
glazing and kinetic facades, but most of the results were not quantitative. Previous findings and
case studies used occupants’ feedback to indicate the thermal and glare performance of EC
glazing and kinetic facades. No obvious glare reduction or thermal improvement was reported by
occupants in the building with the adoption of EC glazing, but there was a slight reduction in
interior blinds usage (Fernandes 2021). Hence, buildings using EC glazing still caused some
thermal or glare issues that impacted occupants. While it was hard to compare the visual and
thermal performance of the combinatorial system to the performance results of EC glazing and
kinetic facades in previous studies, there was plenty of data on energy reduction brought by EC
glazing and kinetic shading.

178
HVAC cooling loads could be reduced by about 30% to 60% in different zones of the building
based on the study done on a physical building (Fernandes 2021). Another study indicated that
EC glazing could reduce 52% energy consumption in cooling-dominated areas (Cannavale
2018). The effectiveness of kinetic shading systems to reduce energy use relied on different
forms and parameters, and studies showed the potential of energy reduction was about 20% to
30% (Hosseini 2019). However, the results of experiment #2 in this research did not show a
significant increase in energy reduction brought by the combinatorial system. The combination
of EC glazing and kinetic miniature blinds reduced 53% peak HVAC cooling load and 26% EUI
from the double-pane glazing option in the Entrada office model. However, the values were not
comparable because the building conditions were different. For the Entrada project, the strategy
of kinetic miniature blinds reduced 45% peak HVAC cooling loads and 16% EUI, and the
strategy of EC glazing reduced 43% peak HVAC cooling load and 24% EUI compared to the
clear double-pane glazing option. The results were different from the previous findings due to
different building conditions, different parameters of EC glazing, and different forms of kinetic
shading. According to this research, the combinatorial system did have better performance in
reducing energy, and thus showed the importance of comparing multiple shading and glazing
strategies in one building setup.
Thus, due to the different composition of the two models, eight test conditions in experiments #1
and #2 had different performance as well. The one having good performance in experiment #1
did not perform well as well in experiment #2. Some performance goals had achieved in
experiment #2 but not in experiment #1.

179
6.1.1 Key Findings
For the shoebox model, test condition #7 of EC glazing and kinetic miniature blinds had the
overall highest points among all test conditions, earning a total of 42 points (Table 5-17). For the
corner room of the Entrada project, test condition #7 earned the overall highest points again,
which were 40 points (Table 5-32).
6.1.1.1 Key Findings of Performance
If daylighting performance was the most important factor to consider, the code-compliant
double-pane glazing without any exterior shading systems provided the most amount of
daylighting for both experiments. For the shoebox model, an average of 24.31% of areas
received sufficient daylighting (Table 5-4), and an average of 62.89% of areas received sufficient
daylight for the Entrada project (Table 5-19). Test condition #8, EC glazing with static overhang
had the least amount of daylight available, which was 6.16% for the shoebox model (Table 5-4)
and 23.23% for the Entrada project (Table 5-19). This might indicate that Spatial Daylight
Autonomy (sDA) might not be the most suitable indicator of daylighting performance, because it
was obvious that clear glazing without any shading system would have most of the daylighting
availability, but it did not mean all of the daylighting was useful. When the illuminance was too
high, it caused visual discomfort, and a large amount of daylight would be a burden. Previous
studies had used Useful Daylight Illuminance (UDI) which indicated the percentage of areas
receiving illuminance from 100 lux to 2000 lux for 50% of the time to evaluate daylight
performance of EC glazing, and EC glazing could improve 82.7% of UDI (Cannavale 2018).
Thus, test condition #1 with clear glazing might have a large percentage of illuminance larger
than 2000 lux that could be disturbing to occupants, and should not be included in the
daylighting performance. Then other test conditions with shading and glazing techniques might

180
have a better performance in daylight due to their potential to reduce time and areas receiving
high illuminance. However, an obvious thing was that receiving more daylight increased the
possibility of glare. Shielding glare would decrease the daylight availability as well.
If glare was the most important issue, EC glazing with kinetic miniature blinds shielded all
disturbing or intolerable glare and provided 12 hours of imperceptible glare for the Entrada
project (Table 5-21). The clear glazing with miniature blinds shielded disturbing and intolerable
glare as well, but it left out 5 hours with perceptible glare (Table 5-21) because the solar
irradiance value was not high enough for the kinetic blinds to be deployed. With clear glazing,
there were an hour of intolerable glare, 2 hours of disturbing glare, and 7 hours of perceptible
glare (Table 5-21). The EC glazing made contributions to the perceptible glare, while the
miniature blinds made contributions to the severe glare. All test conditions had intolerable glare
hours in the shoebox model experiment because the miniature blinds were not deployed at 5 pm
when the glare was the most severe. The severe glare did not equal high solar irradiance.
The shoebox model and the Entrada office model had different results for the rest of the
indicators. EC glazing with kinetic miniature blinds in the shoebox model provided the least
amount of peak cooling load of 1.44 kilowatts (Table 5-7) and Energy Use Intensity (EUI) of
70.39 kWh/m
2
(Table 5-8). The highest value of peak cooling load and EUI provided by clear
double-pane glazing were 2.55 kW (Table 5-7) and 73.86 kWh/m
2
(Table 5-8). However, the
combinatorial strategy of EC glazing with the static overhang used in the Entrada office provided
the least amount of peak cooling load, 12.02 kW (Table 5-22), while the highest peak cooling
load provided by double-pane glazing was 25.34 kW (Table 5-22). 53% of the peak cooling load
was reduced. The combinatorial system of EC glazing and kinetic horizontal blinds used in the
Entrada office had the lowest EUI, which was 68.99 kWh/m
2
, and the highest EUI resulting from

181
double-pane glazing was 93.35 kWh/m
2
(Table 5-23). 26% of the EUI was reduced with the
combinatorial strategy. Nevertheless, the energy use of the Entrada project was only an estimate
based on available information and speculation. The actual building was code-compliant and
energy-efficient. It could have a much smaller EUI and peak cooling load provided by its energy-
saving techniques.
The thermal performance had more discrepancy. The double-pane glazing system of the shoebox
model had the greatest percentage of comfortable conditions, which was 72.07%, in the
Predicted Mean Voted (PMV) model, while EC glazing with kinetic miniature blinds was
comfortable only 56.13% of the time (Table 5-9). However, most of the cold thermal conditions
were caused by the low heating setpoint of the HVAC system, not the shading or glazing
strategies. Therefore, the experiments used the percentage of overheated conditions instead of the
percentage of comfortable conditions as the indicator of PMV thermal performance. The only
problem was that only test condition #1 had 0.64% of the overheated time. The rest of the test
conditions did not have any overheated time, so they all had the tied highest PMV thermal
scores. This was because the window area of experiment #1 was too small, so there was too little
overheated time to solve, and it was hard to compare the performance of those test conditions.
This problem was also reflected in the Predicted Percentage of Dissatisfied (PPD) value and
scores of the shoebox model. While the PPD value larger than 20% was disturbing and lower
than 10% was comfortable, all test conditions had PPD values lower than 20%, and test
conditions #2 to #8 had PPD values lower than 10%. Among them, test condition #7, EC glazing
with kinetic miniature blinds, had the lowest PPD value which was 6.62% (Table 5-10). The
highest PPD value provided by test condition #1, clear double-pane glazing, was 11.93% (Table
5-10).

182
For the Entrada project, the system combining EC glazing and kinetic horizontal blinds had the
most comfortable time (PMV), which was 96.26% (Table 5-24), while the strategy of double-
pane glazing provided the least amount of comfortable time, which was 59.01% (Table 5-24).
However, test conditions #6 to #8 had the best performance in reducing overheated time. Test
condition #7 combining EC glazing with kinetic miniature blinds had zero overheated time
(Table 5-24). Test conditions #6 and #8, which were combinatorial systems of EC glazing with
kinetic horizontal blinds and static overhang, had overheated time below 1% (Table 5-24). Test
condition #1 that had clear double-pane glazing had 36.10% overheated time (Table 5-24), which
was the highest value among all test conditions. Test condition #1 also had the highest PPD
value for the overheated conditions, which was 54.58% (Table 5-25), while test condition #7 had
the lowest PPD value, which was 9.33% (Table 5-25). Therefore, test condition #7 had
successfully maintained the overheated levels within an acceptable range.
6.1.1.2 Analyses of Performance
The discrepancy of results in the two experiments was caused by the window areas. Factors like
daylighting, glare, solar heat, HVAC cooling loads, and overheated thermal conditions were
positively correlated to window areas. Tinted glazing and external shading reduced the factors
mentioned before while increasing HVAC heating loads and cold thermal conditions. Indicators
of EUI, PMV, and PPD considered multiple factors, so the test condition earning the highest
scores for those categories had a good balance among those trade-offs. The rank of peak cooling
load was also different for the shoebox model and the Entrada project. This was related to the
orientation of windows. The shoebox model had a small west-facing window, while the Entrada
room had two large west-facing and south-facing window areas. There was more amount of
sunlight and solar heat coming from the south window all day, but the west window only had a

183
large amount of sunlight and solar heat in the afternoon before the sunset. In this case, the tinted
EC glazing and static overhangs shielding two facades at the same time resulted in less peak
cooling loads than tinted EC glazing with miniature blinds only deployed on one façade.
Initially, results from the PMV thermal comfort conditions and PPD values were unexpected.
The limitation of PMV and PPD metrics was that they were both influenced by the cold time
conditions, so the ability of those shading and glazing techniques to reduce overheated time was
not completely reflected in the results.
While the shading and glazing strategies were used to decrease overheated conditions, they
created a lot of cold conditions as well. Due to small amounts of glazing areas, there were a lot
of cold conditions over a year. Strategies of shading and glazing systems increased more cold
conditions, so the comfortable conditions decreased. For experiment #2, PMV thermal comfort
showed the improvement of comfortable conditions brought by different shading and glazing
techniques because there were not too many cold conditions, but the highest PPD values of the
year indicated the coldest condition in winter mornings, not the overheated conditions.
This was mostly caused by using the default cooling and heating schedule of the large office
building (Table 6-1). The unoccupied hours had different HVAC setpoints to save energy, and
the uncomfortable cold time happened in unoccupied hours. For test condition #4, the lowest
indoor temperature was 16.26 °C, and the highest temperature was 27.80 °C (Table 4-63). The
uncomfortable cold sensation was predicted to be more disturbing than the overheating
sensation. The results did not demonstrate that those shading and glazing techniques increased
cold conditions and created uncomfortable conditions for occupants.

184
Table 6-1 Default Heating and Cooling Setpoints for the Large Office Buildings
HVAC \ Hours 12 am to 5 am 6 am to 9 pm 10 pm to 11 pm
Heating Setpoints 15.6 °C 21 °C 15.6 °C
Cooling Setpoints 26.7 °C 24 °C 26.7 °C
Therefore, the experiments had to focus only on the results of overheated conditions and
excluded the cold thermal conditions to evaluate the thermal performance of different shading
and glazing strategies. The highest PPD value was not an appropriate indicator for thermal
performance. If the most uncomfortable conditions happened during the unoccupied hours, it
would not cause thermal issues to the occupants. Perhaps the real sensations of occupants would
also be different from the simulation results. For future work, thermal data could be collected
from subjects that experienced the shading or glazing techniques in the physical testbed facilities,
and the results could be largely different.
While 80% energy reduction was not achieved by any of the test conditions in either experiment,
sDA, sGA, and indoor temperature range had been achieved by some test conditions in
experiment #2. Test conditions #1, #2, #3, and #5 had sDA on the fall equinox day larger than
55%. They were double-pane glazing, kinetic horizontal blinds, kinetic miniature blinds, and EC
glazing. None of the combinatorial strategies (#6, #7, #8) had met the goal because they were
good at shielding daylighting instead of providing daylighting. Two test conditions with static
overhangs had not achieved the goal either because they were static and shielding the daylighting
all day long.
Test conditions #3 and #7 achieving the goal of 0% sGA40%5%. They both had kinetic miniature
blinds with tilt angles of 45°. This meant only miniature blinds with 45° tilt angles were able to
block the intolerable glare at 5 pm when the sun set. Test condition #3 had more perceptible
glare than test condition #7, meaning EC glazing was able to shield less extreme glare.

185
Test conditions #5 and #6 were able to maintain the annual indoor temperatures within the range
of 19.4 °C to 27.8 °C that was neither too cold nor too hot. This meant EC glazing was good at
keeping temperatures within a certain range. EC glazing with horizontal blinds could still
maintain the temperature in that range, while EC glazing with miniature blinds or overhangs
decreased the temperatures. However, while other test conditions had lower bounds of
temperatures influenced by the heating setpoint 15.6 °C, test conditions #5 to #8 had higher
temperatures that the heating system during the unoccupied hours was probably not used.
All eight test conditions had met the goal of view qualities in both experiments, which indicated
that occupants could still enjoy views while using all those shading or glazing techniques.
Nonetheless, none of the test conditions in either of the experiment was able to reduce 80% EUI
from a regional average. This demonstrated that the intervention on the glazing and facades was
not sufficient to reduce a large amount of energy. More actions should be taken in electric
lighting, equipment, or energy generation by installing solar panels.
Nonetheless, all test conditions being analyzed did not represent the performance of the whole
shading or glazing types. Test condition #5 had better performance compared to test condition #2
did not demonstrate that EC glazing had better performance of kinetic facades. This only
indicated that EC glazing with certain parameters performed better than kinetic blinds with
certain parameters in certain circumstances. A single change of parameters would lead to a
completely different result. Overall, the research provided a method to evaluate and compare the
different performance of various shading and glazing systems.

186
6.2 Future Work
There were subjects, factors, and limitations that had not been covered in the research but could
be included in future work. They were listed here as problems that could be addressed by future
researchers.
6.2.1 Short-term Problems
The Ladybug Tools could simulate solar irradiance data of a year and the incident solar
irradiance for each hour of a day. However, when the annual data was parsed, and the specific
value of an hour of a day was selected from the 8760 data, it was different from the incident solar
irradiance of the hour of that day. The reason was that the components calculating annual solar
irradiance and incident solar irradiance used different methods of calculation. The annual solar
irradiance data was moderately scaled down to match the incident solar irradiance value in this
research. However, they still could not completely match each other. This problem could
probably be solved when the Ladybug Tools got upgraded but should be explored further.
The experiments did not use specific values of solar irradiance to control the dynamic systems.
Experiment #1 used 10%, 20%, 30%, and 40% of the highest annual solar irradiance value
received on the window to switch on EC light, heavy, fully tinted states, and kinetic blinds.
Experiment #2 used 4%, 7%, 11%, and 15% of the highest annual solar irradiance value on the
west façade to switch on dynamic glazing and shading systems on the west façade. 8%, 15%,
23%, and 30% of the highest annual solar irradiance detected on the south façade was used to
initiate dynamic glazing and shading systems on the south façade. The values were not random.
They made sure the annual solar irradiance data and incident solar irradiance value matched, and
the experiments could show all the changing states of dynamic systems on the façades of the
shoebox model and the Entrada office model. If different values were used, the results would be

187
largely different. If a controller other than solar irradiance was used to switch on the dynamic
glazing or shading systems, the results would also be largely different.
The situations of rooms without the HVAC system were only tested for test conditions #1 and #4
in the experiments. Those were clear double-pane glazing and static overhang. Most of the test
conditions were active strategies, and they worked with the HVAC system. If the power outage
happened, only the static overhang could still work. Nonetheless, it would be still useful to test
the thermal conditions when the office building was naturally ventilated without the air-
conditioning. Although the Entrada project did not have operable windows, closed windows
without the HVAC systems could be tested for all test conditions and included in the scoring
system. In that case, adaptive thermal comfort would be used. It was more effective than the
PMV thermal comfort in testing naturally ventilated buildings.
However, this led to another simulation problem that could possibly be solved with the upgrade
of the simulation tools. There was questionable data of thermal comfort when the room was not
air-conditioned and not naturally ventilated when the shading and glazing strategies included EC
glazing. Test condition #5 was to use EC glazing. According to the thermal comfort results of
previous experiments, the highest indoor temperature of test condition #5 without the HVAC
system was expected to be lower than test condition #1 which only had the double-pane glazing.
The expectation aligned with the non-air-conditioned and air-conditioned shoebox model and the
air-conditioned situation of the Entrada project. However, for the non-air-conditioned situation
of the Entrada project, the highest indoor temperature of test condition #5 was 20 °C higher than
test condition #1 (Table 6-2). The reason for this error was unknown but it is speculated that this
will probably be solved with a few updates to the software. If that was true to the real case, then
more research could be done to study the phenomenon. EC glazing may have the potential to

188
gather more heat when there were large window areas. While the simulation results showed that
the Entrada building had extremely high indoor temperatures without the mechanical cooling
system, they were only estimates. The actual building may have a more advanced glazing system
with triple-pane glazing and a lower VLT or other solutions that leads to a lower indoor
temperature. The research only provided estimates and suggestions.
Table 6-2 Adaptive Thermal Comfort of Experiments #1 and #2
Test
Conditio
ns
% Hot %
Neutral
(Comfort
able)
% Cold Lowest
Temperat
ure (°C)
Highest
Temperat
ure (°C)
Temperat
ure
Variation
(°C)
Air-
condition
ed?
Experiment #1 – Shoebox Model
#1 71.83 27.28 0.89 17.98 36.49 18.51 No
#2 69.90 29.21 0.89 17.98 34.73 16.75 No
#3 68.24 30.87 0.89 17.98 33.92 8.75 No
#4 61.23 37.33 1.44 17.56 33.59 8.86 No
#5 63.58 35.64 0.78 18.41 32.65 14.24 No
#6 52.93 45.70 1.37 17.96 31.41 13.45 No
#7 36.06 61.43 2.51 17.44 30.33 12.89 No
#8 56.79 42.10 1.11 18.13 31.62 13.49 No
Experiment #2 – Entrada Project
#1 100 0 0 25.25 74.44 49.19 No
#2 99.66 0.34 0 21.11 51.47 30.36 No
#3 92.40 7.17 0.43 18.10 38.65 20.55 No
#4 99.97 0.03 0 24.09 66.61 42.52 No
#5 100 0 0 35.16 90.46 55.3 No
#6 100 0 0 32.57 78.82 46.25 No
#7 100 0 0 28.12 65.88 37.76 No
#8 100 0 0 33.25 82.93 49.68 No
The Entrada project model did not include furniture in the simulation. This was because the
addition of furniture would make the simulation time much longer. The test conditions with
combinatorial systems already took hours to simulate. Test condition #6 that had EC glazing
with kinetic horizontal blinds took around six hours to simulate the peak cooling loads and more
than two hours to simulate the thermal comfort. Test condition #7 that had EC glazing with more
189
numbers of miniature blinds took an even longer time to simulate due to the number of slats, but
the time was not recorded.
Moreover, there could be more work done on the scoring system. The goal of the scoring system
was to more directly show which of the test condition had superior performance across all
performance indicators and all the trade-offs. However, both of the experiments had test
conditions with tied scores, so several test conditions had the same ranking. In the future, a more
explicit scoring system should be developed to cover this limitation.
6.2.2 Long-term Goals
There were lots of other factors that could be included in the research that were not included due
to limitations of time. The test conditions had horizontal blinds with large spacings between slats
and miniature blinds with small spacings between slats and 45° of slats’ angles. However,
vertical blinds were not included in the test conditions. Vertical blinds could be effective for
windows on the east and west facades. They shielded the sunlight not by altitude but by the
azimuth of the sun (Schiler 2021). The future work could compare the performance of horizontal
blinds and vertical blinds since a lot of office buildings have glass facades on the east and west
sides. One test condition could be having horizontal louver blinds on the south façade and
vertical louver blinds on the west, north, and east facades. The experiments could also have more
controls of the parameters of the test conditions. For instance, there could be blinds with the
same aspect ratio but different tilt angles, or different setpoints of solar irradiance or internal
results.
Another test condition that could be added was to have kinetic blinds that change the tilt angles
of slats. Test conditions #2 and #6 had louver blinds with zero degrees of tilt angles, and test

190
conditions #3 and #7 had 45° of tilt angles. However, their changing states were whether to
deploy them, not to change their tilt angles. To have changeable tilt angles would be very
complicated for the simulations, but could be done with batch runs and it is doable in the reality,
and it may change the results of the performance of kinetic blinds.
Due to the complexity of kinetic blinds and dynamic glazing, the experiments used an approach
to simulate momentary illuminance of each hour only on the fall equinox day and calculated the
average percentage of areas with illuminance larger than 300 lux to indicate the percentage time
receiving sufficient daylighting for all eight test conditions. The fall equinox was used to
represent the whole year because that day received about an average amount of sunlight
compared to other days. However, more data would be more accurate to estimate the daylight
performance. Spatial Daylight Autonomy (sDA) calculated the data for the whole year, but the
values could not be simulated for the kinetic systems. Probably in the future, the sDA results
could be available for the kinetic systems.
If sDA performance was still not available, daylight and glare performance of summer and
winter should be simulated and included in the experiments. The variables of three seasons could
lead to completely different results. If the summer and winter solstice days were also included in
the experiments for the daylight and glare research, the results would be different and more
authentic.
For the EUI calculation, the percentage of areas receiving sufficient daylighting was subtracted
from the energy use of artificial lighting because it compensated for the electrical lighting
energy. However, the percentages receiving sufficient daylight were daily values instead of
annual values, so they were not accurate. They could only be used to estimate the daylight

191
compensation. Moreover, while calculating EUI and subtracting its artificial lighting energy
when illuminance was larger than 300 lux, it did not consider that when the glare was intolerable
or disturbing, the daylight should not be counted, because occupants would deploy the interior
shades to block the glare completely. Hence, the data regarding the amount of daylighting
becomes useless when glare reaches intolerable levels. A more complicated calculating system
should be developed for the EUI calculation.
Moreover, simulations could still not be comparable to the situations in real life. If there were
enough budget and time, a test facility could be built to test the performance of different shading
and glazing options based on physical situations. When the simulation results stated it was too
hot for occupants, the subjects might think it was comfortable. When the simulation results stated
the glare was not disturbing, subjects might find it intolerable. Also, occupants might have
different requirements for daylight and artificial lighting. A change in the physical test facilities
could provide completely different test results from the simulation results. More performance
indicators like costs and carbon footprints would be worth assessing.
View quality was another index that was difficult to evaluate. Take EC glazing as an example, it
changed tinted states while people could still access views through it. However, the view quality
through slightly tinted glazing was different from clear glazing, but it was hard to evaluate how
much did the tinted states affect view qualities. Also, when there was no EC glazing or exterior
shading systems, occupants could not access views when the glare was intolerable. Therefore,
the most accurate assessment of the view quality of the eight test conditions was the reflection
from occupants, which could only be realized with a physical testbed facility.

192
6.3 Conclusion
The research provided several possible circumstances for designers so that they could have a
basic concept of performance in different categories while selecting from those shading and
glazing options. The research also provided methods to evaluate the various performance of
different shading and glazing options. The strategy of “Small Multiples” was used to visually
present the daylight and glare performance, so that it was straightforward to compare, analyze,
evaluate and understand the performance visually through a series of images with the same scale.
The scoring system compared the comprehensive performance of various shading and glazing
options in visual, energy, and thermal categories.
For architects and designers considering applying those shading or glazing strategies to the
buildings, this research may be a useful reference. Test condition #7 earned the highest overall
score in both experiments (Table 5-17 and 5-32), indicating that for either small spaces with
small windows areas or a large office building with large glazing areas, a combinatorial system
with EC glazing and kinetic miniature blinds had an exceeding performance considering visual,
thermal, and energy factors as the hypothesis proposed. Other best performers in each of the
categories were covered before. Overall, in both experiments, strategies with EC glazing (test
conditions #5, #6, #7, #8) performed better than strategies without EC glazing. Among strategies
with double-pane glazing, strategies with external shading (test conditions #2, #3, #4) performed
better than the one without any exterior shading.
Again, the results were only based on certain parameters and certain circumstances. The results
did not mean EC glazing was generally better than double-pane glazing or that kinetic facades
were generally better than static facades. A small nuance in the selection of parameters or in the

193
values obtained could lead to a completely different result in performance. Architects and
designers will need to evaluate and make decisions case by case.

194
REFERENCES
“Bullitt Center.” Bullitt Center, 2013, https://bullittcenter.org/.
“Meeting the 2030 Challenge.” Architecture 2030, architecture2030.org/2030_challenges/2030-
challenge/.
“U.S. Energy Information Administration - Eia - Independent Statistics and Analysis.” Energy
Information Administration (EIA)- Commercial Buildings Energy Consumption Survey
(CBECS) Data. https://www.eia.gov/consumption/commercial/data/2012/.
“WELL Building Standard v1.” The Building Standard - Well V1 Standard | Iwbi.
https://www.wellcertified.com/certification/v1/standard
ANSI/ASHRAE (2017) Standard 55: 2017, Thermal Environmental Conditions for Human
Occupancy. ASHRAE, Atlanta.
Aspen Publisher. “Creator of ‘Radiance’ Simulation Earns Award for Daylight
Research.” Energy Design Update (New York: Aspen Publishers, Inc) 38, no. 6 (2018):
8–9.
Cannavale, Alessandro, Ubaldo Ayr, and Francesco Martellotta. “Innovative Electrochromic
Devices: Energy Savings and Visual Comfort Effects.” Energy procedia 148 (2018):
900–907.
Carbonnier, Kevin, Cathy Higgins, Katie Wilson, Michael Mutmansky, John Rossi, and Paul
Mathew. “Blind to Blinds: Opening Our Eyes to Savings from New Automated Shading
Systems” (2021).
Cheng, Yihong. “Electrochromic Glass Thermal Comfort: Proposals for Improving Existing
Workflow”. ProQuest Dissertations Publishing, 2017.
Fernandes, Luis, Eleanor S Lee, Darryl Dickerhoff, Anothai Thanachareonkit, Taoning Wang,
and Christoph Gehbauer. “Electrochromic Window Demonstration at the John E. Moss
Federal Building, 650 Capitol Mall, Sacramento, California” Lawrence Berkeley National
Laboratory LBNL – 2001183 (2021).

195
Godoy-Shimizu, Daniel, Philip Steadman, Ian Hamilton, Michael Donn, Stephen Evans, Graciela
Moreno, and Homeira Shayesteh. “Energy Use and Height in Office Buildings.” Building
Research & Information 46, no. 8 (2018): 845–63.
Gunter, Bert. "Small Multiples Part 1: Revisiting an Old Friend." Quality Progress, 04, 1997,
129-133, http://libproxy.usc.edu/login?url=https://www.proquest.com/magazines/small-
multiples-part-1-revisiting-old-friend/docview/214520098/se-2?accountid=14749.
Hansanuwat, Ryan. “Kinetic Facades as Environmental Control Systems: Using Kinetic Facades
to Increase Energy Efficiency and Building Performance in Office Buildings”. ProQuest
Dissertations Publishing, 2010.
Herd, David, Heidi Creighton, and Julian Parsley. “Santa Monica City Services Building.” Buro
Happold, April 22, 2021. https://www.burohappold.com/projects/santa-monica-city-
services-building/#.
Hosseini, Seyed Morteza, Masi Mohammadi, Alexander Rosemann, Torsten Schröder, and Jos
Lichtenberg. “A Morphological Approach for Kinetic Façade Design Process to Improve
Visual and Thermal Comfort: Review.” Building and Environment 153 (2019): 186–204.
Jones, Nathaniel L. “Fast Climate-Based Glare Analysis and Spatial Mapping.” Building
Simulation Conference proceedings, September 2, 2019.
Koh, Kelly, Hussain H. Al-Kayiem, and Jundika C. Kurnia. “Thermal Comfort Assessment of an
Office Building in Tropical Climate Condition.” MATEC Web of Conferences 225
(2018): 01003.
Ko, Won Hee, Michael G. Kent, Stefano Schiavon, Brendon Levitt, and Giovanni Betti. “A
Window View Quality Assessment Framework.” LEUKOS, 2021, 1–26.
Lee, Dong-Seok, Sung-Han Koo, Yoon-Bok Seong, and Jae-Hun Jo. 2016. "Evaluating Thermal
and Lighting Energy Performance of Shading Devices on Kinetic Façades" Sustainability
8, no. 9: 883.
Lee, Eleanor, Luis Fernandes, Samir Touzani, Anothai Thanachareonkit, Xiufeng Pang, and
Darryl Dickerhoff. “Electrochromic Window Demonstration at the 911 Federal Building,
911 Northeast 11th Avenue, Portland, Oregon” Lawrence Berkeley National Laboratory
LBNL – 2001195 (2016).

196
LM, I. "Approved method: IES spatial Daylight autonomy (sDA) and annual sunlight exposure
(ASE)." Illuminating Engineering Society. https://www. ies. org/product/ies-spatial-
daylight-autonomy-sda-and-annual-sunlight-exposure-ase (2013).
Long, Nicholas, Eric Bonnema, Kristin Field, and Paul Torcellini. “Evaluation of
ANSI/ASHRAE/USGBC/IES STANDARD 189.1-2009.” National Renewable Energy
Laboratory, 2010. https://doi.org/10.2172/984670.
Love, J. A., W. Tian, and Z. Tian. “Window-to-Wall Ratios and Commercial Building
Environmental Control in Cold Climates.” CiteSeerX. University of Calgary, 2019.
https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.667.2389.
LPC West. “Floor Plates.” ENTRADA, March 8, 2021. https://www.entrada-la.com/entrada-
floor-plates.
Luo, Kunyu. “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”. ProQuest Dissertations Publishing, 2018.
Mackey, Chris. “Has Anyone Did a Simulation of Electrochromic Window?” Ladybug Tools |
Forum, Jan. 10, 2019, discourse.ladybug.tools/t/has-anyone-did-a-simulation-of-
electrochromic-window/4891/6.
Mardaljevic, J, and A Nabil. “Electrochromic Glazing and Facade Photovoltaic Panels: a
Strategic Assessment of the Potential Energy Benefits.” Lighting Research & Technology
(London, England: 2001) 40, no. 1 (2008): 55–76.
Nicol, J.F., and M.A. Humphreys. “Adaptive Thermal Comfort and Sustainable Thermal
Standards for Buildings.” Energy and Buildings 34, no. 6 (2002): 563–72.
O'Brien, William, Konstantinos Kapsis, and Andreas K. Athienitis. “Manually-Operated
Window Shade Patterns in Office Buildings: A Critical Review.” Building and
Environment 60 (2013): 319–38.
Paule, Bernard, Eloise Sok, Samuel Pantet, and Julien Boutiller. “Electrochromic Glazings:
Dynamic Simulation of Both Daylight and Thermal Performance.” Energy Procedia 122
(2017): 199–204.

197
Pittaluga, M. “Electrochromic Glazing and Walls for Reducing Building Cooling Needs.”
In Eco-Efficient Materials for Mitigating Building Cooling Needs: Design, Properties
and Applications, 474–497, 2015.
Sayadi, Sana, Abolfazl Hayati, and Mazyar Salmanzadeh. “Optimization of Window-to-Wall
Ratio for Buildings Located in Different Climates: An Ida-Indoor Climate and Energy
Simulation Study.” Energies, 1974, 14, no. 7 (2021): 1–21.
Sbar, Neil L., Lou Podbelski, Hong Mo Yang, and Brad Pease. “Electrochromic Dynamic
Windows for Office Buildings.” International Journal of Sustainable Built Environment
1, no. 1 (2012): 125–39.
Schiler, Marc. “Shading Devices, Shading Masks, and Critical Angles.” Class lecture, Design for
the Luminous and Sonic Environment from University of Southern California, Los
Angeles, CA, September 30, 2021.
Smith-Gardiner, Nina. “The Curtain Wall System at the Bullitt Center.” Living proof blog.
Bullitt Center, July 18, 2012. https://bullittcenter.org/2012/07/18/the-curtain-wall-
system-at-the-bullitt-center/.
Suk, Jae Yong & Schiler, Marc, & Kensek, Karen. “DISCOMFORT GLARE METRICS:
Investigating their accuracy and consistency in daylight glare evaluation by using human
subject study data.” Facade Tectonics World Congress. Los Angeles, 2016
Wienold, J, T Iwata, M Sarey Khanie, E Erell, E Kaftan, RG Rodriguez, JA Yamin Garreton, et
al. “Cross-Validation and Robustness of Daylight Glare Metrics.” Lighting Research &
Technology (London, England: 2001) 51, no. 7 (2019): 983–1013.
Wienold, Jan and Christoffersen, Jens. “Towards a New Daylight Glare Rating.” Lux Europa.
Berlin, 2005
Williams, Brian. “Photochromic Thermochromic and Electrochromic Glass - Climate Change.”
Brian Williams, July 29, 2019, www.briangwilliams.us/climage-change-2/photochromic-
thermochromic-and-electrochromic-glass.html.
Woodford, Chris. “How Do Electrochromic (Automated Glass) Windows Work?”
Electrochromic Windows, Apr. 23, 2011, www.explainthatstuff.com/electrochromic-
windows.html.

198
Vincent, Roger. “Plush Indoor-Outdoor Office Building to RISE near Playa Vista and 405
Freeway.” Los Angeles Times. Los Angeles Times, December 13, 2018.
https://www.latimes.com/business/la-fi-lincoln-entrada-20181210-story.html.

Abstract (if available)

Abstract Conventional static glazing sometimes has poor performance in energy, visual, and thermal aspects. In this thesis, a series of simulations of an office building were done to compare the performance of conventional static glazing, exterior static and kinetic shades, dynamic glazing, and dynamic glazing working together with static or kinetic shades as a combinatorial system. Energy uses include annual Energy Use Intensity (EUI) and peak cooling loads of the hottest day of the year. Visual performance includes maximizing daylight and minimizing glare. Thermal performance includes Predicted Mean Vote (PMV) thermal comfort that was based on occupants’ sensation of thermal conditions and Predicted Percentage of Dissatisfied (PPD) which indicated the levels of thermal discomfort. A scoring system was developed to evaluate the overall performance. The prediction was that the combination of dynamic glazing and kinetic shade would have the best comprehensive performance among visual, energy, and thermal comfort.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

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

PDF

Microclimate and building energy performance

PDF

Real-time simulation-based feedback on carbon impacts for user-engaged temperature management

PDF

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

PDF

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

PDF

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

PDF

Occupant-aware energy management: energy saving and comfort outcomes achievable through application of cooling setpoint adjustments

PDF

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

PDF

Modular shading: using design to mitigate bus rider thermal heat stress

PDF

The effectiveness of enviro-materially actuated kinetic facades: evaluating the thermal performance of thermo-bimetal shading component geometries

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

Developing environmental controls using a data-driven approach for enhancing environmental comfort and energy performance

PDF

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

PDF

Solar thermal cooling and heating: a year-round thermal comfort strategy using a hybrid solar absorption chiller and hydronic heating scheme

PDF

Tiny house in the desert: a study in indoor comfort using moveable insulation and thermal storage

PDF

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

PDF

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

PDF

Mitigating thermal bridging in ventilated rainscreen envelope construction: Methods to reduce thermal transfer in net-zero envelope optimization

PDF

Building energy performance estimation approach: facade visual information-driven benchmark performance model

Asset Metadata

Creator Lu, Weixuan (author)

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

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Degree Conferral Date 2022-05

Publication Date 04/13/2022

Defense Date 04/12/2022

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

Tag daylight,electrochromic glass,Energy,glare,grasshopper,kinetic shades,ladybug tools,OAI-PMH Harvest,thermal comfort

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

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

Creator Email weixuan@usc.edu,weixuanlu9@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC110937308

Unique identifier UC110937308

Document Type Thesis

Format application/pdf (imt)

Rights Lu, Weixuan

Type texts

Source 20220413-usctheses-batch-923 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu