University of Southern California Dissertations and Theses

"Watts per person" paradigm to design net zero energy buildings: examining technology interventions and integrating occupant feedback to reduce plug loads in a commercial building

(USC Thesis Other)

"Watts per person" paradigm to design net zero energy buildings: examining technology interventions and integrating occupant feedback to reduce plug loads in a commercial building

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content "WATTS PER PERSON” PARADIGM TO DESIGN NET ZERO ENERGY
BUILDINGS:
EXAMINING TECHNOLOGY INTERVENTIONS AND INTEGRATING
OCCUPANT FEEDBACK TO REDUCE PLUG LOADS IN A COMMERCIAL
BUILDING
By

Mika Yagi Kim

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

August 2013

Copyright 2013 Mika Yagi Kim
ii

Acknowledgements
This thesis is important to me. I believe that the content of the research investigated
here can make a difference in the AEC community toward a new way of designing and
delivering Net Zero Energy Buildings. In the process of writing this thesis, I came across
many challenges with the subject matter as it is not yet come to the forefront of research
topics. With that said, I was fortunate to be supported by individuals who believed in my
mission to demystify one of the most unaccounted energy end uses in buildings today -
plug loads in commercial buildings.
First and foremost, I thank my newly wedded husband, Mark Kim, for his patience and
understanding throughout this process. With his support and encouragement, I was able
to work countless of hours to meet all the deadlines in the writing schedule. His
persistence in ensuring that my work was being backed up helped to relieve my anxiety
more than words can describe after one incident when I failed to do so and lost all of my
work. Thanks also to my family and friends who have been supportive of my decision to
pursue my Masters degree much later in life.
I would next like to thank my committee members, Kyle Konis and Marc Schiler from
USC School of Architecture and Ken Hall from Gensler. All were instrumental in the
development of the this thesis. Thank you, Kyle for helping me get familiarized on
academic terminology and providing me input from that perspective. I am grateful for
your time during our weekly meetings where you helped to provide me direction when it
was needed. I have added vocabulary in my language that now includes interventions,
implications and study design. Thank you, Marc for being Uncle Marc. When times got
iii

difficult, you were the person that helped me navigate my way through the process.
Thank you, Ken for believing that my work is valuable and important to the field of
Architecture. I look forward to collaborating with you on future projects at Gensler.
Much gratitude is given to the members who supported my work from the peripheral,
Jeff Landreth from CTL-E, Michael Sheppy from the National Renewable Energy
Laboratory (NREL), and Doug Noble from USC School of Architecture for your guidance,
technical advice and encouragement. Thank you, Jeff for showing me how excel should
really be used. My Chapter 6 has been elevated due to your influence and overall
participation helped to strengthen this thesis. Thank you, Michael for your
encouragement throughout the process. The correspondences we exchanged throughout
the year with your feedback gave me the reassurance I needed that my work was relevant
and important. When there were times that no one seemed to understand what I was
doing, you gave me the confidence that I was on the right track. I admire the work you
have done for NREL and hope we can collaborate on future research together. Thank
you, Doug for establishing a high caliber for the program this year.
Finally, I thank the other talented eleven members in the MBS program who on a week-
to-week helped me to feel normal for the extreme amounts of stress and anxiety we all
experienced to meet each writing schedule deadline. Everyone in this program has a
bright future and I look forward to working on a project together in some capacity after
this.

iv

Table of Contents
Acknowledgements .............................................................................................................. .. ii
List of Tables ......................................................................................................................... iv
List of Figures ....................................................................................................................... vi
Abstract .................................................................................................................................. 1
Chapter 1 – Statement of Intent ............................................................................................ 2
1.1. Hypothesis .............................................................................................................. 2
1.2. Policy Framework ................................................................................................... 2
1.3. International Policy ................................................................................................ 4
1.4. Standards and Guidelines ...................................................................................... 5
1.4.1. ASHRAE .................................................................................................................. 5
1.4.2. ASHRAE Handbook ............................................................................................... 7
1.4.3. ASHRAE 50% Advanced Energy Design Guide .................................................... 8
1.5. Leadership in Energy & Environmental Design .................................................... 8
1.6. Plug Load Impacts on Net Zero Energy Buildings .............................................. 10
1.6.1. Case Study Building ............................................................................................... 11
1.7. Current Methods for Estimating Plug Loads ....................................................... 12
1.7.1. Watts Per Person Benchmark Approach ............................................................. 13
Chapter 2 – Background Data ............................................................................................. 15
2.1. Energy Use Tracking Summary ........................................................................... 16
2.2. Equipment Audit Summary .................................................................................. 17
2.2.1. Load Intensity Defined ......................................................................................... 19
2.2.2. Load Intensity Classification Summary ............................................................... 21
2.2.3. Equipment Audit Using Nameplate Data ............................................................ 21
2.3. Power Modes ........................................................................................................ 23
2.4. Power Management Settings................................................................................ 24
2.4.1. High Performance ................................................................................................ 24
2.4.2. Balanced ............................................................................................................... . 24
2.4.3. Power Savings ....................................................................................................... 24
2.4.4. Custom .................................................................................................................. 25
v

2.4.5. No Control ............................................................................................................ 25
2.5. Parasitic Loads ...................................................................................................... 26
2.6. Watts Per Person Analysis ................................................................................... 27
Chapter 3 – Research .......................................................................................................... 28
3.1. Research Goals ..................................................................................................... 28
3.1.1. Technology to Reduce Energy .............................................................................. 28
3.1.2. Feedback to Influence Behavior ........................................................................... 30
3.2. Methodology ......................................................................................................... 31
3.2.1. No Control ............................................................................................................ 32
3.2.2. Baseline ................................................................................................................. 32
3.2.3. Control ................................................................................................................ .. 33
3.2.4. Daily Feedback ...................................................................................................... 34
3.3. Hypothesis: Part 1 (Advanced Controls) .............................................................. 34
3.3.1. Load Sensing Control ........................................................................................... 35
3.3.2. Occupancy Sensor Control ................................................................................... 35
3.3.3. Remote Switch Control......................................................................................... 36
3.3.4. Timer Schedule Control ....................................................................................... 36
3.3.5. APS Summary for Reference ................................................................................ 37
3.4. Hypothesis: Part 2 (Daily Feedback) ................................................................... 38
3.4.1. Watts Per Person Feedback .................................................................................. 38
3.4.2. Daily Feedback Notification ................................................................................. 38
3.4.3. Competition as a Factor ....................................................................................... 39
Chapter 4 – Research Data Review ..................................................................................... 40
4.1. Measurement Results ........................................................................................... 40
4.1.1. Metered Data ........................................................................................................ 41
4.2. Load Profile Analysis ............................................................................................ 42
4.2.1. Heavy Load User Profile ....................................................................................... 43
4.2.2. Medium Load User Profile ................................................................................... 47
4.2.3. Light Load User Profile ........................................................................................ 51
4.2.4. Ultra Light Load User Profile ............................................................................... 55
4.3. PL Energy Use Summary ..................................................................................... 59
vi

4.3.1. PL Energy Use Summary for Reference .............................................................. 60
4.4. Lessons Learned ................................................................................................... 63
4.4.1. People as Subject Matter ...................................................................................... 63
4.4.2. Understanding of Equipment Operability ........................................................... 64
4.4.3. Unregulated = Not Highly Funded for Research ................................................ 64
Chapter 5 – Inferences from Data ...................................................................................... 65
5.1. Modeled (Estimate) PLs vs. Measured (Actual) PL Energy Use Overview ........ 65
5.1.1. Heavy Load User Profile ....................................................................................... 67
5.1.2. Medium Load User Profile ................................................................................... 70
5.1.3. Light Load User Profile ........................................................................................ 73
5.1.4. Ultra Light Load User Profile ............................................................................... 76
5.2. Modeled vs. Measured Comparative Analysis ..................................................... 79
5.3. Summary of Analysis ........................................................................................... 80
5.4. Load Factor Analysis ............................................................................................ 83
5.5. Lessons Learned ................................................................................................... 84
5.5.1. Use Standard Load Profiles to Estimate Standard Load Factors Only ............... 84
5.5.2. Assumption of Standard Load Profiles ................................................................ 84
Chapter 6 - Miscellaneous Load Overview ......................................................................... 86
6.1. Appliances ............................................................................................................. 86
6.1.1. Microwave ............................................................................................................. 87
6.1.2. Refrigerator ........................................................................................................... 88
6.1.3. Coffee Maker ......................................................................................................... 89
6.1.4. Ice Maker .............................................................................................................. 90
6.1.5. Filtered Water Dispenser ..................................................................................... 91
6.1.6. Cappuccino Maker ................................................................................................ 92
6.2. Imaging Equipment ............................................................................................. 93
6.2.1. Laser Printers ....................................................................................................... 94
6.2.2. Laser Copiers ........................................................................................................ 95
6.2.3. Inkjet Printers ....................................................................................................... 96
6.3. Audio Visual Equipment ...................................................................................... 97
6.3.1. 42" LED/LCD Flat Screen Display ....................................................................... 98
vii

6.4. Miscellaneous Load Analysis ............................................................................... 99
6.5. Controls Evaluation ............................................................................................ 100
6.6. Lessons Learned ................................................................................................. 102
6.6.1. Loads Most Appropriate for Controls ................................................................ 102
6.6.2. Rated (W) ≠ Peak (W) ........................................................................................ 102
6.6.3. Specify Energy Star Rated Equipment .............................................................. 102
Chapter 7 – Conclusion ..................................................................................................... 104
7.1. Key Findings ....................................................................................................... 105
7.2. Standard Load Profiles Re-Examined ............................................................... 106
Chapter 8 – Future Work .................................................................................................. 109
8.1. Analysis of Other Building Types ....................................................................... 109
8.2. Increase Sampling per Building Type ................................................................ 109
8.3. Evaluate Load Profiles per Advanced Controls ................................................. 109
8.4. Advanced Controls for Miscellaneous Loads ......................................................110
8.5. Incentives as a Factor for Further Reduction .....................................................110
8.6. Plug Loads as a Priority in Building Design ......................................................110
Bibliography: ................................................................................................................. ..... 112
Appendix ...................................................................................................................... ....... 114
Applicable Rating Systems ................................................................................................. 114
Intent......................................................................................................................... .......... 114
Requirements .................................................................................................................. ... 114
Performance ................................................................................................................... .... 116
Credit Submittals ................................................................................................................ 117
Credit Specific: .............................................................................................................. ...... 118
Additional Questions .......................................................................................................... 118
Background Information .................................................................................................... 118
Appendix-Data ................................................................................................................. ... 119
iv

List of Tables
Table 1.2: California’s Roadmap to Energy Efficiency (CPUC, 2008)....………......…………..3
Table 1.4.1: Code Compliance on Receptacle Loads…………………........……........………….…..6
Table 1.4.2: Recommended Equipment Load Factors..……………………………........……...…..7
Table 1.4.3: AEDG Recommendation Table.............……………........……………………….……..8
Table 1.5: Default Tenant Receptacle Loads by Occupancy Type......................................10
Table 1.7: Plug Load Factor Benchmark........…………………………........…….....................….13
Table 2.2.1 : Equipment Use per Load Profile.................……………………………...........……20
Table 2.2.3: Computing Equipment Audit...............………………………….........……………….22
Table 2.3: Power Mode Summary....................................……………………..........……..………25
Table 2.3.5: Power Management Setting Evaluation....……………………….......………….……20
Table 2.6: Plug Load Factor Benchmark based on Watts Per Person………........……….....27
Table 3.2: Pilot Study Schedule...........................................................................................31
Table 3.1.2: Daily Feedback Protocol.............................……………………...........….………….30
Table 3.2.1: No Control, Power Management Setting......………………………….......………….32
Table 3.2.2.: Recommended Control, Power Management Setting………………........………33
Table 3.2.3: Power Saver Control, Power Management Setting…………………….........…….33
v

Table 3.2.4: Custom Control, Power Management Setting……..………….………………......…34
Table 3.2: APS Controls per Load Intensity Classification…………………………………....... 35
Table 3.3.4: APS Control Evaluation, Advantages & Disadvantages….....……….........…….37
Table 4.3.1: Plug Load Factor based on Watts per Person................................................60
Table 4.3.1.1: Heavy Load Profile, Average Watt Hours per Person……….….………..........61
Table 4.3.1.2: Medium Load Profile, Average Watt Hours per Person..............................61
Table 4.3.1.3: Light Load Profile, Average Watt Hours per Person ……………….....…..……62
Table 4.3.1.4: Ultra Light Load Profile, Average Watt Hours per Person.........................62
Table 5.2: Comparative Analysis of Modeled vs. Measured Deviation..............................79
Table 6.4: Rated Power vs. Peak Watts .............................................................................99
Table 6.5: Recommended Controls for Typical Office Equipment...................................101
Table 7.1: Pilot Study Implications Summary..................................................................105
Table A1- Appendix: Plug Load Reduction, LEED Pilot Credit.......................................100
vi

List of Figures
Figure 1.3.1.-Global Rate of Energy Consumption Based on Watts Per Person..................5
Figure 1.6 – Electricity Growth Projections (EIA, 2008)…..……………..…………….....……...11
Figure 1.6.1– RSF Energy Use Breakdown (NREL, 2010).…………………………..…….…….12
Figure 2.1– Energy Use Tracking Summary (By Month)...................................................17
Figure 2.2-PL Energy Use Estimate Comparison...............................................................18
Figure 2.2.2– Load Intensity Classification Summary ..............……………………….............21
Figure 4.2.1.1– Heavy Load Profile, No Control….............................................................45
Figure 4.2.1.2– Heavy Load Profile, Baseline………………….....………………….….…..…….….45
Figure 4.2.1.3 – Heavy Load Profile, Load Sensing Control & Power Mgmt…..…….....….46
Figure 4.2.1.4– Heavy Load Profile, Control with Daily Feedback.......................…….……46
Figure 4.2.2.1- Medium Load Profile, No Control.........................................................….49
Figure 4.2.2.2- Medium Load Profile, Baseline.................................................................49
Figure 4.2.2.3- Medium Load Profile, Occupancy Sensor & Power Mgmt........................50
Figure 4.2.2.4- Medium Load Profile, Control with Daily Feedback.................................50
Figure 4.2.3.1- Light Load Profile, No Control...................................................................53
Figure 4.2.3.2- Light Load Profile, Baseline.......................................................................53
vii

Figure 4.2.3.3- Light Load Profile, Remote Switch & Power Mgmt..................................54
Figure 4.2.3.4- Light Load Profile, Control with Daily Feedback......................................54
Figure 4.2.4.1- Ultra Light Load Profile, No Control.........................................................57
Figure 4.2.4.2- Ultra Light Load Profile, Baseline.............................................................57
Figure 4.2.4.3- Ultra Light Load Profile, Timer Control & Power Mgmt..........................58
Figure 4.2.4.4- Ultra Light Load Profile, Control with Daily Feedback............................58
Figure 5.1–Typical Computational Load Profiles for Commercial Buildings....................59
Figure 5.1.1.1–Heavy Load Profile, No Control..................................................................68
Figure 5.1.1.2–Heavy Load Profile, Baseline......................................................................68
Figure 5.1.1.3–Heavy Load Profile, Load Sensing Control & Power Mgmt.......................69
Figure 5.1.1.4–Heavy Load Profile, Control with Daily Feedback.....................................69
Figure 5.1.2.1–Medium Load Profile, No Control..............................................................71
Figure 5.1.2.2–Medium Load Profile, Baseline..................................................................71
Figure 5.1.2.3–Medium Load Profile, Load Sensing Control & Power Mgmt...................72
Figure 5.1.2.4–Medium Load Profile, Control with Daily Feedback.................................72
Figure 5.1.3.1–Light Load Profile, No Control...................................................................74
Figure 5.1.3.2–Light Load Profile, Baseline.......................................................................74
Figure 5.1.3.3–Light Load Profile, Load Sensing Control & Power Mgmt........................75
viii

Figure 5.1.3.4–Light Load Profile, Control with Daily Feedback......................................75
Figure 5.1.4.1–Ultra Light Load Profile, No Control..........................................................77
Figure 5.1.4.2–Ultra Light Load Profile, Baseline..............................................................77
Figure 5.1.4.3–Ultra Light Load Profile, Timer Control & Power Mgmt...........................78
Figure 5.1.4.4–Ultra Light Load Profile, Control with Daily Feedback.............................78
Figure 5.3- Summary of Load Factor Deviation................................................................80
Figure 5.3.1–Heavy Load Profile, Measured vs. Modeled Analysis...................................81
Figure 5.3.2–Medium Load Profile, Measured vs. Modeled Analysis...............................81
Figure 5.3.3–Light Load Profile, Measured vs. Modeled Analysis....................................82
Figure 5.3.4–Ultra Light Load Profile, Measured vs. Modeled Analysis..........................82
Figure 5.4-Load Factor Analysis.........................................................................................83
Figure 6.1.1- Microwave, Average Watt Hours...................................................................87
Figure 6.1.2- Refrigerator, Average Watt Hours................................................................88
Figure 6.1.3- Coffee Maker, Average Watt Hours..............................................................89
Figure 6.1.4- Ice Maker, Average Watt Hours....................................................................90
Figure 6.1.5- Filtered Water Dispenser, Average Watt Hours............................................91
Figure 6.1.6- Cappuccino Maker, Average Watt Hours.....................................................92
Figure 6.2.1- Shared Laser Printers, Average Watt Hours.................................................94
ix

Figure 6.2.2- Shared Multi Function Laser Copier, Average Watt Hours.........................95
Figure 6.2.3- Shared Inkjet Oversize Printer, Average Watt Hours..................................96
Figure 6.3.1- 42” LED/LCD Flat Screen Display, Average Watt Hours............................98
Figure 6.4- Miscellaneous Load Summary.........................................................................99
Figure 7.2.1- Heavy Load Profile, Load Sensing Control & Power Mgmt........................107
Figure 7.2.2– Medium Load Profile, Control with Daily Feedback.................................107
Figure 7.2.3- Light Load Profile, Remote Switch Control & Power Mgmt......................108
Figure 7.2.4- Ultra Light Load Profile, Timer Control & Power Mgmt............................108
Figure A1 Appendix- Heavy Load Profile, No Control.....................................................119
Figure A2 Appendix- Heavy Load Profile, Baseline.........................................................119
Figure A3 Appendix- Heavy Load Profile, Control...........................................................120
Figure A4 Appendix- Heavy Load Profile, Daily Feedback..............................................120
Figure A5 Appendix- Medium Load Profile, No Control..................................................121
Figure A6 Appendix- Medium Load Profile, Baseline......................................................121
Figure A7 Appendix- Medium Load Profile, Control.......................................................122
Figure A8 Appendix- Medium Load Profile, Daily Feedback..........................................122
Figure A9 Appendix- Light Load Profile, No Control.......................................................123
Figure A10 Appendix- Light Load Profile, Baseline.........................................................123
x

Figure A11 Appendix- Light Load Profile, Control...........................................................124
Figure A12 Appendix- Light Load Profile, Daily Feedback..............................................124
Figure A13 Appendix- Ultra Light Load Profile, No Control............................................125
Figure A14 Appendix- Ultra Light Load Profile, Baseline................................................125
Figure A15 Appendix- Ultra Light Load Profile, Control.................................................126
Figure A16 Appendix- Ultra Light Load Profile, Daily Feedback.....................................126
1

Abstract
As building envelopes have improved due to more restrictive energy codes, internal loads
have increased largely due to the proliferation of computers, electronics, appliances,
imaging and audio visual equipment that continues to grow in commercial buildings. As
the dependency on the internet for information and data transfer increases, the
electricity demand will pose a challenge to design and operate Net Zero Energy Buildings
(NZEBs). Plug Loads (PLs) as a proportion of the building load has become the largest
non-regulated building energy load and represents the third highest electricity end-use
in California's commercial office buildings, accounting for 23% of the total building
electricity consumption (Ecova 2011,2). In the Annual Energy Outlook 2008
(AEO2008), prepared by the Energy Information Administration (EIA) that presents
long-term projections of energy supply and demand through 2030 states that office
equipment and personal computers are the "fastest growing electrical end uses" in the
commercial sector.
This thesis entitled “Watts Per Person" Paradigm to Design Net Zero Energy Buildings,
measures the implementation of advanced controls and behavioral interventions to
study the reduction of PL energy use in the commercial sector. By integrating real world
data extracted from an energy efficient commercial building of its energy use, the results
produce a new methodology on estimating PL energy use by calculating based on "Watts
Per Person" and analyzes computational simulation methods to design NZEBs.
2

Chapter 1 – Statement of Intent
This chapter is an introduction to the thesis topic and explains: 1) policy and framework
of PL use in commercial buildings, 2) impact of PLs in NZEBs, and 3) current industry
standard assumptions to calculate PL energy use. The "Watts Per Person” paradigm is
explained as the unit measure to benchmark PL standards for NZEBs.
1.1. Hypothesis
Two (2) hypotheses were developed in this pilot study to address two (2) PL
interventions.
Hypothesis 1: Technology to Reduce Energy
Implementation of Advanced Plug Strips (APS) will reduce PL energy use
Hypothesis 2: Feedback to Influence Behavior
When given the objective to reduce PL energy use, occupants given daily feedback on
their PL energy use reduce more PL use than occupants without feedback
1.2. Policy Framework
In October 2007, the California Public Utilities Commission (CPUC) adopted the Long
Term Energy Efficiency Strategic Plan to set forth a roadmap for energy efficiency in
California through 2020 and beyond, in which one of the four programmatic goals in the
"Big Bold Energy Efficiency Strategies" is for new and a substantial proportion of
existing commercial buildings in California to be put on a path to Net Zero Energy
3

Building (NZEB) by 2030. At present, commercial buildings consume more electricity
than any other end-use sector in California and of that, office buildings are the largest
consumers (CPUC 2008, 30).

Table 1.2. - California's roadmap for energy efficiency through 2020 and beyond
(Source: CPUC)

Due to the advancement of building technology in envelope, lighting and HVAC systems,
the overall energy use has reduced in the commercial sector while the dependency on
computers and other electronic office equipment has expanded, increasing in its demand
load (see Figure 1.6). In achieving NZEBs, end users will need to reduce the PL demand
to minimize on-site energy generation. The implementation of both technology and
change in occupant behavior will be necessary to reduce PL energy use in buildings and
furthermore, research is needed to understand how to better predict PL energy use
through computational simulation in the design phase.
4

1.3. International Policy
In 2008, the City of Zurich made a ground breaking decision and adopted the goal of the
"2,000-Watt Society" into its municipal code developed by researchers at the Swiss
Federal Institute of Technology (ETH), as the model for energy policy to demonstrate its
commitment to sustainable development. The energy policy is based on every person in
every society limiting their energy consumption to a maximum of 2,000-Watts per
Person with at least 75% of its source to be met by renewable energy. Although the
energy use fluctuates from country to country, 2,000-Watts is the average energy
consumption per capita world-wide and takes into account of all aspects of sustainability
from energy demand in buildings to mobility, food, consumption and infrastructure (City
of Zurich 2011, 7). The PL energy use demand is addressed within the building category
where it targets its limit to 500-Watts per Person.
In the United States, the figure is six times higher at 12,000-Watts per Person while
countries such as Bangladesh, Zimbabwe and rural areas of China need barely a fraction
of the average energy consumption. In 1985, research conducted by Brazilian scientist,
Jose Goldemberg found that below a threshold of 1,000-Watts per Person, people are
indeed better off if they can increase their energy consumption, however, once this
threshold is reached, more energy does not improve the quality of life. The goal of the
2,000-Watt Society enables a balance between the industrialized and developing
countries use of energy to address concerns of Global Climate Change but more
importantly, is a tool that allows the industrialized countries to be informed on its
energy use status per person in comparison to the worldwide average and a reminder
that it requires everyone to get involved. This concept of Watts per Person will be
5

adopted for this thesis as a metric tool to inform building occupants on its energy use per
person.

Table 1.3.1. - Global rate of energy consumption based on Watts Per Person (Source: PSI)
1.4. Standards and Guidelines
At present, PLs are classified as non-regulated energy and are addressed in the building
code standards only in terms of controls. Supplemental guidelines and modeling
requirements are available for calculating proposed and baseline building performance
to design teams which are discussed in the following sub-sections in this chapter.
1.4.1. ASHRAE
American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE)
90.1 is the basis for the International Energy Conservation Code (IECC) and is widely
adopted as the minimum energy efficiency standard for most states. The most recent
6

version of the standard was published in 2010 and in Section 8.4.2 , it is stated that 50%
of receptacles are required to be on automatic shutoff systems, however the load itself is
not being regulated. Generally, regulated energy includes heating, ventilation, air
conditioning, water heating, and lighting.
In the reference standard, there is no definition that currently exists of plug or receptacle
loads but in its Appendix section, there are guidelines that exist of modeling
requirements for calculating proposed and baseline building performance for receptacle
and other loads (see Section 1.5 for more information).

Automatic Receptacle Control

ASHRAE
90.1-2010

ASHRAE
90.1-2007

IECC
2012

CA Title
24/2008

Automatic shutoff of 50% of all 125 volt
receptacles in private offices, open
offices, and computer classrooms
(including receptacles installed in
modular partitions through:
X

1) Scheduled Shutoff

X

2) Occupancy Sensor (shutdown within
30 minutes)

X

3) Signal from another control/alarm
system
X

Exceptions: 1) receptacles dedicated to
equipment with 24 hour operation, 2)
spaces where automatic shutoff would
endanger safety/security of occupants

X

Table 1.4.1. - Code Compliance on Receptacle Controls
7

1.4.2. ASHRAE Handbook
The 2009 ASHRAE Handbook - Fundamentals, include basic principles and data used
in the HVAC&R industry as guidelines that cover general information to estimate PL
density. In Chapter 18 of the Nonresidential Cooling and Heating Load Calculations,
equipment load factors are included as recommendations based on breakdowns as
shown in Table 1.3.2. The general guidelines were developed by ASHRAE Research
Project RP-822 with the intention to develop a method to measure the heat gain from
equipment in buildings. In the recommended Power Density, the breakdown is shown to
include general office equipment and excludes computer peripherals, server equipment,
appliances, telephones, audio visual and task lighting.

Recommended Load Factors for Various Types of Offices

Load Density of
Office

Load Factor
(W/ft²)

Description of Equipment Types

Light

0.50

(6) Computers, (6) Monitors, (1) Laser Printer, (1) Fax
Machine

Medium

1.00

(8) Computers, (8) Monitors, (1) Laser Printer, (1) Fax
Machine

Medium/Heavy

1.50

(10) Computers, (10) Monitors, (1) Laser Printer, (1)
Fax Machine

Heavy

2.00

(12) Computers, (12) Monitors, (1) Laser Printer, (1)
Fax Machine
Table 1.4.2. - Recommended Equipment Power Load Factor (Source: Wilkins and Hosni,
2000)
8

1.4.3. ASHRAE 50% Advanced Energy Design Guide
The 2011 Advanced Energy Design Guide (AEDG) for Small to Medium Office Buildings
developed by ASHRAE, AIA, IESNA, USGBC and U.S. DOE provides design strategies
and recommendations to achieve 50% energy savings over the minimum code
requirements of ANSI/ASHRAE/IESNA Standard 90.1, 2004. PLs reductions are
addressed through the following recommendations:

Plug Load Reduction Recommendations

Equipment
Choices

Laptop computers

Minimum 2/3 of total computers

ENERGY STAR
equipment

For all computers, equipment, and appliances

Equipment Power
Density

For all computers, equipment, and appliances

Controls

Computer Power
Control

Network control with power savings modes
and control OFF during unoccupied hours

Occupancy Sensors

Desk plug strip occupancy sensors

Timer Switches

Water coolers and coffee makers control OFF
during unoccupied hours

Vending Machine
Control

Yes
Table 1.4.3. - AEDG Recommendation Table
1.5. Leadership in Energy & Environmental Design
The 2009 Edition of the Reference Guide used for LEED® green building certification
programs defines plug loads to be synonymous with receptacle loads by identifying it as
9

current drawn by all equipment that is plugged into the electrical system. LEED
establishes one of the minimum level of energy efficiency by addressing new eligible
equipment to be ENERGY STAR qualified by 50% (by rated power) and includes
appliances, office equipment, electronics, and commercial food service equipment. In
terms of parameters within the whole building simulation method, LEED relies entirely
on the performance rating method in ASHRAE 90.1-2007, Appendix G for both the
baseline building model and the proposed building model to cover all building energy
components. In the Appendix G, PLs are included in the process energy load estimation
which includes office and general miscellaneous equipment, computers, elevators and
escalators, data center and telecom room computing equipment, kitchen cooking and
refrigeration, laundry washing and drying, lighting exempt from lighting power
allowance. In the modeling requirements for the calculation, the baseline case for the
total plug and process loads are required to be identical to the proposed building and
cost must be equal to at least 25% of baseline building performance. To model the
proposed building, values in Table 1.5 are provided as default plug loads. These values
are mentioned that they do not necessarily reflect all process loads but the values are
recommended to achieve the 25% process loads.
10

Table 1.5. - Default Tenant Receptacle Loads, by Occupancy Type (Source: LEED)
1.6. Plug Load Impacts on Net Zero Energy Buildings
PLs are miscellaneous building energy loads that are not direct HVAC, building
envelope, process, lighting or domestic hot water loads and are heavily owner, operator,
occupant driven demands. In energy efficient buildings, PLs can contribute up to 50% or
more of the total Energy Use Intensity (Lobato 2010). With NZEBs, the baseline
represents a value of zero and the term "net" implies, in most cases, that the building will
use the utility grid as its "battery" source to charge when the building produces more
energy than it consumes and draws at other times when it consumes more than it is
producing. Thus, to operate a building that consumes no more energy than it produces
(and potentially produces more energy than it uses for a net positive outcome) poses a
challenge to examine reduction measures of all energy use including those that are non-
regulated. The U.S. Energy Information Administration (EIA) Annual Energy Outlook
forecasts PL demands to increase 94% from 2005 to 2030 (see Figure 1.6.) which will
classify its demand as one the major contributors in energy use in commercial buildings.
11

Figure 1.6. - Electricity Demand Load Growth Projections (Source: EIA 2008 Annual
Energy Outlook)

1.6.1. Case Study Building
The importance of PL energy use reduction is strongly emphasized by the award winning
Department of Energy's National Renewable Energy Laboratory Research Support
Facility (RSF) project completed in 2010. By adopting a net zero energy approach to
construct its large scale commercial building, the energy use intensity (EUI) totaled to
35.4 kBtu/ ft²/yr, with the PL energy use contributing to over 25% of the total EUI and
found to be the second highest energy demand next to process loads. To achieve
maximum PL energy use reductions, the project used the finding that for every one (1)
Watt (W) of continuous energy use saved; $33 of Photovoltaic (PV) System cost that
would be required to offset this energy saved. This case study project is a demonstration
of how important the consideration of PL energy use is in high performance buildings.
12

Figure 1.6.1. - RSF Energy Use (kBtu/ft²) Breakdown (Source: NREL, 2010)

1.7. Current Methods for Estimating Plug Loads
Based on current industry standards, PLs are typically estimated by using a combination
of the following variables: 1) specified area per workstations or offices (for example, 85
ft²/Workstation), 2) quantity of equipment type within the specified area (for example, 6
computers/6 monitors), 3) peak electrical load (W), 4) load factor (W/ft²), 5) design max
occupancy (for example, 200 ft²/Person), and 5) usage diversity to determine the ratio of
measured peak electrical load to the sum of the maximum electrical load of each
individual item of equipment. These variables in conjunction with consideration of time
using load profiles defined by building types are often factors to determine PL
consumptions in buildings.
13

Based on research funded by the California Energy Commission (CEC) Public Interest
Energy Research Program (PIER) performed by the New Building Institute (NBI), the
published PL Performance Level guidelines suggests that load factors for PLs can be as
low as 0.25 W/ft². These plug load factors will be used in the pilot study to set the
benchmark.

Plug Load Factor Benchmark (W/ft²)

Poor

Standard

High
Performance

Best

Occupied Load
Factor

0.75+

0.75

0.40

0.25*

Unoccupied Load
Factor

0.70+

0.70

0.35

0.20**

Combined Load
Factor

1.45+

1.45

0.75

0.45

Peak Demand
Density

1.5+

1.5

1.0

0.75*

* Includes Load Factor suggested by Wilkins,Hosni, "Plug Load Design Factors" ASHRAE Journal in May 2011 that
includes equipment of: 100% Laptop Use, (1) Printer per 10, Speakers & Miscellaneous per 167 ft²
**Load Factor based on NREL's RSF case study
Table 1.7. – Plug Load Factor Benchmark (NBI & PIER Analysis)
1.7.1. Watts Per Person Benchmark Approach
Technology in equipment continually evolve to become more efficient and trends in
space planning continually evolve, where private offices today are being converted to
open plan workstations to optimize real estate value and as a result, it has made it
challenging to estimate PL energy use based on Watts per Area (ft²). Due to these
variables that are not always constant, the current method of estimating PLs is flawed in
14

that it fails to address these changing trends, including the growing demand of shared
spaces in buildings and planning strategies that include unassigned seating to encourage
mobility that require more use of laptop equipment by users.
To take these factors into account when estimating PL energy use, this thesis suggests
that power density assumptions (W) and area (sf²) need to be replaced with more focus
on the load intensity classifications by user types (determined by power consumption of
the equipment) and the average load (W) per Person over time to determine "Watts per
Person", adopted by the concept used for the 2,000-Watt Society described in the
preceding section. This method allows for customization to estimate PL energy use that
take into account of changing trends of equipment type usage and its space use. In
Chapter 2, further explanations of the load intensity classifications are described in
detail as the basis of the Watts per Person calculation method.

15

Chapter 2 – Background Data
This chapter provides background data of the pilot study by explaining: 1) the origin of
the real world data of the energy use summary for the high performance commercial
building, 2) summary of the equipment audit and descriptions of each load intensity
classification, and 3) definitions to related terms and standards. In the last section, the
Watts per Person metric used as the benchmark for the pilot study is further outlined in
detail.
As mentioned in the previous chapter, PLs in high performance buildings can exceed
50% of its total energy use. This chapter evaluates the energy use consumption for a
commercial building completed in 2011 awarded LEED Platinum certification and
analyzes the results by focusing on PLs, the highest electricity load measured for the
building with the highest discrepancy between the modeled and measured outcome.
The three-story 40,000 ft² project was designed to house over 300 occupants of 30 ft²
per Person allocated for the open plan benching desk system concept with multiple
shared spaces throughout the building. In addition to the multiple strategies used to
optimize energy use, PL reduction strategies were implemented by replacing the needs
of individual task lighting with integrated task ambient lighting system throughout and
specifying 100% of its new equipment to be ENERGY STAR rated.
Submeters were installed in the building to separately monitor the energy use for
lighting systems/controls, plug loads, process loads, heating and cooling which were all
being monitored through the Building Automation System (BAS). The actual measured
energy use from the submeters will be shared in the next section.
16

2.1. Energy Use Tracking Summary
While working in conjunction with the design teams, the monthly energy use tracking
summary was developed in order to compare the design phase energy use estimates to
the actual building energy use. By extracting the data from the eQuest energy model
software used during the design phase, energy use estimates were established as the
baseline case to compare to the building energy performance tracked through the BAS.
As a result, six months into reporting, a trend revealed that the highest electricity use
(kWh) consumed in the building came from PL energy use (see Figure 2.1) . In addition,
results indicated that the modeled (estimate) use for PLs were highly underestimated
compared to its measured (actual) use by over 50%. The opposite was true for the
process loads where the server room equipment and all of its associated dedicated
cooling for the equipment indicated that the modeled (estimate) use were highly
overestimated compared to its measured (actual) use by over 50%. Given these results,
this thesis and pilot study will focus on PL energy use from the perspectives of how it can
be reduced and also to further investigate where the discrepancies may have come from
in terms of modeled outcomes compared to actual use.
The understanding of these factors will become critical when designing and operating
NZEBs. The discrepancies between the modeled (estimate) and measured (actual) data
is an issue since with NZEBs, the renewable energy systems may not have been properly
sized in this case to generate these types of unanticipated energy load for the building.
17

Figure 2.1. - Energy Use Tracking Summary (By Month)

2.2. Equipment Audit Summary
As with most organizations and businesses, classifications of office equipment use by
occupants vary based on job descriptions to suit the specific needs. Based on that, it is
nearly impossible to make accurate estimations of PL energy use based on one
standardized load factor of W/ft², which is in most cases the methodology used. In
doing so, results can be demonstrated in Figure 2.2.A where the baseline comparison
utilizes the commonly used load factor of 1.0 W/ft² for Medium Load Density of offices
with the default load profile and 200 ft² per Person (see Figure 5.1. for more
information) used in eQuest energy models while comparing it with the actual people
density of the benching system at 30 ft²/Person (5 ft Wide x 6 ft Deep). The assumption
18

of estimating PL energy use by a standardized load factor and people density was likely
the explanation of the underestimated modeled PL energy use for the building which in
the case of designing NZEBs will become a critical issue.

Figure 2.2. - PL Energy Use Estimate Comparison (with default 1.0 W/ft² and 0.46 load
factor as baseline)

It is also common practice to have multiple classifications of equipment use within the
same building type, which adds another layer of complexity. By conducting the
equipment audit for the energy efficient building, results indicated that four (4) load
intensity classifications were identified and which is described in detail in Section 2.2.1.

19

Due to these factors of having multiple load intensity classifications within one building
type, this pilot study uses the different classifications and categorizes the variations of its
equipment types using four (4) categories: 1) Heavy, 2) Medium, 3) Light and 4) Ultra
Light.
Having these multiple load intensity classifications within one building type when
tabulating the PL energy use helps to prevent from the “one size fits all” calculation
methodology when estimating PL energy use for buildings.
2.2.1. Load Intensity Defined
The breakdown of equipment use by the load intensity classifications discussed in the
previous section is shown in Table 2.2.1. to identify the multiple variables.
These classifications are characterized as follows: 1) Heavy Load - describes user types
that spend < 20% of the time away from the desktop computers throughout the occupied
demand hours and has the highest baseline of W/Person with advanced computing
equipment features to support high processing and memory capacities coupled with two
(2) LED/LCD Flat Panel monitors to enhance productivity for graphic intensive
workflows, 2) Medium Load - describes user types that spend <20% of the time away
from the desktop computers throughout the occupied demand hours and has the second
highest baseline of W/Person with semi-advanced computing equipment features to
support medium processing and memory capacities coupled with one (1) LED/LCD Flat
Panel monitor to perform daily administrative tasks, 3) Light and 4) Ultra Light Load -
describes user types that spend >20% of the time away from laptop computers to
support the high mobility work flow performance and is separated by classification by
20

the monitors with Light as the same monitor as the Medium Load Intensity of (1)
LED/LCD Flat Panel, whereas the Ultra Light Load Intensity is supported by (1)
USB/LED/LCD Flat Panel powered by the laptop computer.

Load Intensity
Classification

Equipment Type

Heavy

Medium

Light

Ultra Light

*Telephones were excluded from the pilot study
Table 2.2.1. - Equipment Use per Load Intensity Classification

21

2.2.2. Load Intensity Classification Summary
Of the total equipment audited (N=1,238) for the pilot study, findings indicated that of
the 302 occupants, 35% were classified as Light and Ultra Light load intensity and the
remaining 48% were Heavy and 17% were Medium load intensity. In summary, desktop
computers comprise 65% of the load intensity classification for the energy efficient
commercial building.

Figure 2.2.2. - Load Intensity Classification Summary

2.2.3. Equipment Audit Using Nameplate Data
The National Electrical Manufacturers Association (NEMA) defines the nameplate data
on equipment to evaluate the basic design and dimensional parameters to identify the
performance of the equipment. The rated power provided on the nameplate refers to the
maximum active power that the device can output under the rated operation conditions
and differs from the active power. The information on the nameplate typically include,
22

but not limited to: 1) Manufacturer's Type, 2) Frequency, 3) Rated Load Current, 4)
Voltage, 5) Ampere and manufacturers often include additional information such as
Input Wattage to further define its features. Since the nameplate data is different than
the active power consumption, calculations of determining PL energy use can be complex
and inconsistent across equipment and user types. There are multiple variables to
consider which include but are not limited to peak load (or peak demand), operating
hours, load (or capacity, or part-load) factor, number of occupants, equipment type and
control type. In later chapters, results of the average active power (W) is presented and
compared against the nameplate data.

PC Equipment List

Load Profile
Classification

# of Users

Quantity

Description

Rated Power (W)

Heavy

144

432

- Desktop Computer
- Dual LED/LCD Monitors

562
144

Medium

52

104

- Mini Tower Desktop
Computer
- LED/LCD Monitor

200
72

Light

105

210

- Laptop Computer
- Docking Station
- LED/LCD Monitor

98
130
72

Ultra Light

1

2

- Laptop Computer
- USB LED/LCD Monitor

65
10

TOTAL 302 748*

*The remaining equipment audited are non-computing equipment and is further discussed in Chapter 6
Table 2.2.3. - Computing Equipment Audit
23

2.3. Power Modes
Most computing, audio visual and imaging equipment have various power modes when
active vs. inactive that are separate from the power management settings that will be
discussed in detail in Section 2.4. The definitions for these modes are as follows:

Power Modes

Level of Power
Consumption

Type

Description

Highest

Peak

- Level of power use: Max

High

Active

- Level of power use: < Peak
- Performing at is intended function

High/Medium

Idle

- Level of power use: < Active
- Ready to perform work but not doing anything

Low

Sleep*

- Level of power use: < Idle
- Wakes up in seconds
- Same as equipment being on "pause"

Low

Hibernates

- Level of power use: < Sleep
- Wakes up in 20+ seconds
- Saves work in the event of power loss
- Primarily used for laptops but available on
desktops

Lower

Standby / Shut
Down (Off)

- Level of power use: < Hibernate
- Turns on in 20+ seconds
- Referred to as Parasitic Load

Lowest

Disconnected /
Unplugged

- Level of power use: zero

*Default system settings for most equipment
Table 2.3. - Power Mode Summary (Source: ENERGY STAR)
24

2.4. Power Management Settings
Power management controls are system settings that manage how computing equipment
uses power. The settings vary by the amount of power consumed and is available in the
following control plans: 1) High Performance, 2) Power Savings, 3) Balanced, and 4)
Custom.
2.4.1. High Performance
The "High Performance" power management control maximizes the screen brightness
and uses more energy and reduces the amount of time that the laptop battery lasts
between charges.
2.4.2. Balanced
The "Balanced" power management control is the system recommended setting and is
the combination of "High Performance" and "Power Savings" and offers full performance
when needed and saves power during periods of inactivity but uses more energy than
"Power Savings".
2.4.3. Power Savings
The "Power Savings" power management control reduces the system performance and
screen brightness and helps the laptop users get the most from a single battery charge.
25

2.4.4. Custom
The "Custom" power management control allows users to modify settings from the
existing default settings. In this option, the sleep and display settings can be customized
to the user's preference including the brightness of the display to optimize performance
or to save more energy.
2.4.5. No Control
The "No Control" power management control disables the system settings and sets the
equipment to be in active mode for 24/7 unless the user's command to shut down. This
setting disables the equipment to go into sleep mode unless it is shut down.

Power Management Settings

Level of Power
Consumption

Type

Description

High

No Control

- Level of power use: > High Performance,
Balanced, Power Savings

High/Medium

High
Performance

- Level of power use: < No Control, > Balanced &
Power Savings

Medium

Balanced

- Level of power use: < No Control & High
Performance, > Power Savings
- Wakes up in seconds

Low

Power Savings

- Level of power use: < High Performance &
Balanced

26

Varies

Custom

- Level of power use: < / > / = Power Savings

Table 2.3.5. - Power Management Setting Summary
2.5. Parasitic Loads
In 1999, the International Energy Agency (IEA) launched the One-Watt Initiative, an
energy savings proposal targeted to reduce parasitic loads in equipment to be one (1)
watt. Many regions and countries such as the European Union, the United States,
Australia, Japan and South Korea have adopted this benchmark as policy. Parasitic
loads are continuous power consumed when equipment is shut down, also referred to as
standby power (often called vampire draw or phantom loads).
According to IEA, parasitic loads are responsible for 1% of the global CO2 emissions, or
about 240 million tonnes of CO2 every year. Most common culprits that contribute to
parasitic loads include devices with LED display or functions that consume power while
they are in their "off" state. Typical examples of devices that consume parasitic loads are
microwaves that are not in use but consumes power from their digital clock/timer
display that stays on 24/7 while the equipment is connected to the power source, devices
with "instant-on" functions such as laptops, computers and monitors, motion sensor
operated equipment, or audio-visual equipment that is operated by remote controls such
as televisions.
Product labels from ENERGY STAR identify products which meet strict energy efficiency
guidelines set by the U.S. Department of Energy (DOE) and on average are 25% more
efficient than standard models (www.energystar.gov). At present, California Title 20
Appliance Efficiency Regulations regulates the external power supply standby power.
27

Parasitic loads become critical considerations when designing NZEBs as each watt
consumed can lead to the need of upsizing the renewable energy system for buildings.
2.6. Watts Per Person Analysis
Based on the W/ft² load factor benchmark reviewed in earlier sections (see Table 1.4.2),
direct conversions to Watts per Person are shown in Table 2.6 which will be used in the
pilot study to benchmark the results. Both occupied and unoccupied demand times will
be evaluated as both are important factors in estimating plug load energy use. The goal
for the pilot study will be to achieve an average of 60 Watts per Person and not higher
than an average of 190 Watts per Person with the interventions being implemented. The
results of the pilot study are presented in future chapters.

Average Watts/Person* (W)

Poor

Standard

High
Performance

Best

Occupied Demand

100+

100

55

35

Unoccupied Demand

90+

90

45

25

Combined Demand

190+

190

100

60

Peak Demand
Density

200+

200

135

100

* Load Factor is calculated based on values provided in Table 1.5. and is based on 40,000 ft² with 302 persons
Table 2.6: Plug Load Factor Benchmark based on W/Person
28

Chapter 3 – Research
This chapter describes the pilot study designed for this thesis that consists of two (2)
parts to address the two hypotheses: 1) to explore the technology interventions by
implementing advanced controls to reduce PL energy and 2) to explore the behavioral
interventions by implementing a daily feedback system given the objective to reduce PL
energy use.
Test subjects in the pilot study were selected based their load intensity classification with
the focus on the devices used by the participants (N=20) to measure the energy use
consumption for a one week duration per intervention. In this chapter, the highest
priority was put on computers, computer peripherals and monitor(s). The data and
analysis can be found in Chapters 4 and 5.
3.1. Research Goals
Below are the research goals that this thesis will attempt to answer based on the two
hypotheses stated in Chapter 1.
3.1.1. Technology to Reduce Energy
Comparison of Rated Power (W) vs. Peak Demand Power (W)
As described in Chapter 2 (see Section 2.2.3.), the rated power from the nameplate data
is often considered as the maximum or peak load for the equipment energy use and those
will be compared against the actual measured peak demand power to assess
discrepancies, if any.
29

Impacts of Power Management Settings
In order to understand the impacts of the power management settings on equipment
performance, a variable for "No Control" (see Section 2.4.5) that disabled all power
management and intervention controls was included to be able to compare those
measured data to the "Baseline".
Baseline PL Energy Performance
To determine the baseline for the pilot study, test subjects selected for the study were
provided explicit instructions to conduct "business-as-usual" and were advised not to
modify any of the power settings on the devices that were being tested. Although there
were variables within those parameters of laptops and desktop computers having
different default settings, network settings on the devices were verified which were all on
the same “balanced" power management settings (see Section 2.4.2). With no other
information given other than that the study measured the test subject’s PL energy use,
the pilot study deployed for one week. As mentioned in the previous section, the study
placed the highest priority on computers, computer peripherals (i.e. cell phone charger)
and monitor(s) per Table 2.2.1.
Implementation of Advanced Controls
To measure the impacts of advanced controls on PL energy use, Advanced Power Strips
(APS) were implemented in the pilot study to assess the efficiency and effectiveness of
the technology intervention while integrating custom power management settings to
reduce energy use which is further detailed in Section 3.3.
30

3.1.2. Feedback to Influence Behavior
To implement the behavioral change intervention using the daily feedback response to
measure PL energy use, test subjects were given information regarding their PL energy
use from the "Controls" results (see Table 3.1.2.) prior to starting the pilot study as a
reference. Each morning for the duration of the week study, test subjects were notified
by email with the previous workday PL energy use to determine whether that would be a
motivation factor that influenced behavior. Although there was no reward for the load
intensity classification with the highest % reduction, a daily winner was identified to
study whether competition was a factor that influenced behavior.

Table 3.1.2. - Daily Feedback Protocol
31

3.2. Methodology
To implement the pilot study to test the interventions to reduce PL energy use,
participants in the four (4) load intensity classifications described in preceding chapters
were given four (4) variables in the study design, each measured for one week duration
that lasted a total of three months. The variables included the following: Variable 1) No
Control, Variable 2) Baseline, Variable 3) Control, and Variable 4) Daily Feedback. The
test subjects were voluntary participants selected based on load intensity classifications,
each representing a different load user profile: 1) Heavy, 2) Medium, 3) Light and 4)
Ultra Light. The selected test subjects remained constant throughout the pilot study to
maintain consistency with the equipment measured. All results were evaluated using
average load (W) per person over time (h).

Pilot Study Schedule*
Schedule Intervention Type Description
Week 1 Technology No Control
Week 2 Technology Baseline
Week 3 Technology Control (Power Management Settings &
Advanced Controls)
Week 4 Behavioral Daily Feedback
*Each intervention measured for each load intensity classification
Table 3.2. - Pilot Study Schedule
32

3.2.1. No Control
The "No Control" test condition is intended to measure the computing equipment that is
disabled of all controls. These settings are often disabled by the users themselves as a
preference to eliminate settings that allow computers and displays to go into "sleep"
mode after a specified time frame.

Table 3.2.1. - No Control, Power Management Setting
3.2.2. Baseline
The "Baseline" test condition is intended to measure the computing equipment to
perform as "business-as-usual" and have the power management settings to the network
recommended settings. The "Baseline" test condition can be modified to have custom
settings and often are adjusted to shorten or extend the specified time frame to allow
computers and displays to go into "sleep" mode.
33

Table 3.2.2. – Recommended Control, Power Management Setting
3.2.3. Control
The "Control" test condition is intended to measure the computing equipment by
implementing advanced controls for optimal performance with the power management
settings on power savers mode. Similarly to the "Baseline" test condition, these can be
modified to have custom settings and often are adjusted to shorten or extend the
specified time frame to allow computers and displays to go into "sleep" mode.

Table 3.2.3. - Power Saver Control, Power Management Setting
34

3.2.4. Daily Feedback
The "Daily Feedback" test condition is intended to measure the computing equipment by
implementing daily feedback protocols to notify users of their energy consumption to
measure whether it has an influence on the behavior. This test condition includes
advanced controls implemented in the "Controls" variable with the option to customize
the power management setting to adjust to shorten or extend the specified time frame to
allow computers and displays to go into "sleep" mode.

Table 3.2.4. - Custom Control, Power Management Setting

3.3. Hypothesis: Part 1 (Advanced Controls)
To evaluate the PL energy use, highest priorities were put on computers, peripherals and
monitor(s) with the implementation of advanced controls to measure the optimal energy
performance. The selected four (4) Advanced Plug Strips (APS) were paired with the
equipment from the four (4) load intensity classifications (see Table 2.2.1) based on
applicability and operability of systems. As an outcome, none of the desktop computers
were controlled due to its inability to automatically shut down.
35

Heavy

Medium

Light

Ultra Light

Load Sensing

Occupancy Sensor Remote Switch

Timer Schedule

Table 3.2. – APS Controls per Load Intensity Classification
3.3.1. Load Sensing Control
The load sensing control has a microprocessor inside that determines when the one (1)
"master" receptacle is on or off to send responses to the other five (5) "master-
controlled" peripheral outlets to shut down. In addition to the six (6) receptacles, there
are two (2) uncontrolled outlets to allow for equipment that requires 24/7 operability.
3.3.2. Occupancy Sensor Control
The occupancy sensor consists of six (6) controlled and two (2) uncontrolled that are
always on for equipment that requires 24/7 use. The controlled receptacles are on
personal occupancy sensors that utilizes passive infrared (PIR) technology to sense
occupancy to turn on and off controlled receptacles. An adjustable time delay of thirty
(30) seconds to thirty (30) minutes for the six (6) "sensor controlled" receptacles. The
36

multi-level lens used on the occupancy detection sensor has 120° coverage and can be
modified to allow for a more narrow detection as needed.
3.3.3. Remote Switch Control
The remote switch control allows for the whole computer system along with its
peripherals to be shut down including the standby power with a single click on a wireless
remote switch. The remote control has a range of up to sixty (60) feet and can be
operated without having a direct line of sight. Six (6) of the eight (8) are controlled on
the automatic shut down switch and the other two (2) uncontrolled that are always on for
equipment that requires 24/7 use.
3.3.4. Timer Schedule Control
The timer schedule allows for control from the moment you require power by clicking on
a button to begin its timer for eleven (11) hours with manual override for the six (6)
controlled receptacles. In the situation that power consumption is not required for the
full eleven (11) hours, users can click on the button to shut-down including the standby
power and in contrast, in the situation that power consumption is required beyond the
full eleven (11) hours, the LED on the control button will begin to click to notify the user
that the power is about to be shut down. In that case, since it is on a manual override,
the user just needs to push the button to continue its power source for its needed
duration. The other two (2) uncontrolled that are always on for equipment that requires
24/7 use.
37

3.3.5. APS Summary for Reference
Below is the table to summarize the characteristics of the advanced controls used for the
pilot study:

APS Evaluation

Spec

Type
Parasitic Load (W) Automatic
Shut Off (Y/N)
On Off

Belkin Conserve Switch

Remote Switch

0.0

0.0

N
Pros:
- Allows ease and convenience to shut down multiple equipment with one switch and
consumes no standby parasitic loads when turned off
Cons:
- On manual shut off which means if user does not switch off, the devices connected to the
controls will not automatically shut down the equipment

Belkin Conserve with
Timer

Timer Schedule

0.0

0.0

Y
Pros:
- On automatic shut off which means if user does not switch off, the devices connected to
the controls will automatically shut down the equipment
Cons:
- Inability to modify the hours on the timer to customize to each user profile

Belkin Conserve AV

Load Sensing

0.0

0.0

N
Pros:
- It senses when the equipment in the master outlet is shut down to automatically shut
down all other computer peripherals connected to shut down along with it
Cons:
- On manual shut off which means if user does not switch off the equipment connected to
the master outlet, the devices connected to the controls will not automatically
shut down the equipment
Isole IDP-3050 Power
Strip with Personal
Sensor
Occupancy
Sensor

2.8 / 2.9

0.0

Y
Pros:
- On automatic shut off which means if user does not switch off, the devices connected to
the controls will automatically shut down the equipment
Cons:
- Constant power consumption during standby mode
38

Table 3.3.4. - APS Evaluation, Advantages & Disadvantages

3.4. Hypothesis: Part 2 (Daily Feedback)
To evaluate the PL energy use, highest priorities were put on computers, peripherals and
monitor(s) with the implementation of feedback response to measure the optimal energy
performance. The feedback intervention uses the primary approach of informing the
selected participants with daily feedback of their energy use consumption for a week
duration following the technology intervention implementation to examine any
behavioral changes.
3.4.1. Watts Per Person Feedback
Each day of the Daily Feedback intervention was structured so that data was compiled
from its previous 24-hour period to calculate the W/Person, which was described in
previous chapters as the average load (W) per person over time (h). The data was
provided each morning prior to the participant’s use of the equipment and notified via
email of the outcome with the results from the previous week shared as reference.
3.4.2. Daily Feedback Notification
The daily results of the daily feedback of W/Person was shared amongst all of the
participants in the pilot study and communicated through email with full transparency
to examine whether knowing information of other’s energy use has an influence on
behavior.
39

3.4.3. Competition as a Factor
To further evaluate the influence of daily feedback as a factor to reduce PL energy use,
results were quantified per day to reveal percentage (%) reduction for each load intensity
classification (see Figures 4.2.1. to 4.2.4) based on the measured outcome. Although
there was no reward for the load intensity classification with the highest % of reduction,
by having a ranking structure in place that identified a winner allowed for the study to
measure whether competition as a factor influenced behavior. This information was
shared in combination with the daily email feedback notification and results will be
discussed in Chapter 4.

40

Chapter 4 – Research Data Review
This chapter presents the data collected from the pilot study to support the hypotheses
as follows: 1) technology to reduce PL energy use and 2) feedback to influence behavior.
The results from the pilots study are presented by providing the following data: 1) the
output results, 2) comparing the measured average W/Person against the baseline, and
3) analyzing the technology and feedback response intervention results.
For the purpose of this thesis, the PL energy use is calculated based on the average power
consumption (W) per person during a typical workday (h). The study evaluates an entire
week of measurement over the full 24 hour period of each day with the focus on the
equipment usage patterns from Monday through Friday to study its load profile by
starting with results based on one minute intervals.
In the figures presented in this chapter, results are summarized by the load intensity
classifications (Heavy, Medium, Light, and Ultra Light) and by intervention type (No
Control, Baseline, Control, and Daily Feedback). The pilot study places the highest
priority on computers, computer peripherals and monitor(s) but results from the other
miscellaneous equipment loads measured can be found in Chapter 6.
4.1. Measurement Results
In this chapter, the real time logged measurements extracted from Wattsup Pro ES will
be presented with results from the four (4) load intensity classifications described in the
preceding chapters categorized based by load consumption : 1) Heavy, 2) Medium, 3)
Light, and 4) Ultra Light.
41

The measured results are presented in load profile figures that represent data from
measurements taken at 1 minute intervals. The loads are presented in Watts (W).
Furthermore, the results are presented in a series of figures for representing each of the
four (4) intervention categories: 1) No Control, 2) Baseline, 3) Control, and 4) Daily
Feedback. These results are summarized in Section4.3 by comparing the percent (%)
reduction for each category.
All findings from Chapters 4 through 6 are used to outline a list of strategies for reducing
PL energy use and can be found in the Appendix of this thesis.
4.1.1. Metered Data
The metered data was recorded for 7 parameters, including: 1) Current Watts, 2) Volts,
3) Amps, 4) Cumulative Watt Hours, 5) Average Monthly kWh, 6) Cumulative Cost, and
7) Power Cycle that recorded for any power interruptions. For the purposes of this
thesis, we will review the exported data measured using Watts (W) by importing it to an
Excel file for further analysis to calculate the average load of the five (5) working days
from Monday through Friday with weekends shown as reference only.
The metered data results of the average watts consumption categorized by the various
load intensity classifications can be found in the figure below (See figures 4.3.1.1 to
4.3.4.4). The load intensity classifications varies by the equipment types used as
described in the preceding chapter and the data summary can be found in Section 4.3.
The heavy load intensity indicates the equipment type with the highest measured output
and the ultra light intensity indicates the equipment type with the lowest measured
output (see Sections 4.2.1. thru 4.2.4. for additional information). The figures
42

summarized in this Chapter are informative to evaluate equipment use patterns and are
used to study the peak demand usage of electricity consumption. This type of data would
be difficult to view and to interpret using minute-by-minute data tabulations so it has
been converted into figures to evaluate the results.
4.2. Load Profile Analysis
As discussed in the preceding chapters, below is the short summary of the variables used
in the pilot study. All the participants representing each load intensity classification
remained constant throughout to have a better understanding of the changes in PL
energy use based on the same equipment measured.
The first variant with the figure denoted as “No Control” is the recorded meter data for
the four (4) load intensity classification types with disabled controls and without the
implementation of advanced controls.
The second variant with the figure denoted as "Baseline" is the recorded meter data for
the four (4) load intensity classification types with enabled power management controls
but without the implementation of advanced controls.
The third variant with the figure denoted as "Control" is the recorded meter data for all
the four (4) load intensity load classification types with enabled power management
controls with the implementation of advanced controls.
The fourth variant with the figure denoted as "Daily Feedback" is the recorded meter
data for the four (4) load intensity load classification types with enabled power
management controls with the implementation of advanced controls, given daily
feedback of their energy consumption.
43

4.2.1. Heavy Load User Profile
The Heavy Load Profile has the highest measured peak value and highest overall energy
use per person. The output is not as high as the rated power on the nameplate data and
at its highest, measured to be more than a third less. The parasitic load, on the other
hand, is the highest amongst the Medium, Light and Ultra Light Load Profiles.
The output data for the Heavy Load Profile can be seen in the figures below. They are
labeled according to the variables and shows the recorded data using average Watts (W).
In the following chapter, we will further analyze these load profiles by evaluating the
average daily load profile, which is calculated based on the average load in watts over
fifteen (15) minute intervals. This average load profile will also be used to compare
against typical commercial building load profiles to better understand the deviation
between the measured data and typical assumptions.
The data output during the "No Controls" variant shows a constant demand requirement
over the occupied and unoccupied demand, where the average watts consumption is 176
W/Person. The "Baseline" variant where the power management is enabled shows a
reduction but shows the most significant impact of load reduction with the use of APS
with power management as the variant where the calculated consumption is 67
W/Person. The "Daily Feedback" has a reduction compared to the "Baseline" measure
but shows a moderate increase compared to the "APS with power management" due to
the increase in its occupied demand. Based on the output data, the highest reduction
achieved in the study case for the Heavy Load Profile is 34%.
The power consumption during the unoccupied demand is the least with the
implementation of the APS, which is nearly twice the occupied hours. Due to the
44

building automation system set up that resets devices at 2am each day to allow for
systems updates to occur during unoccupied hours, it disables the function afterwards
for the equipment to completely shut down and results in continuous consumption of
power thereafter.

45

Figure 4.2.1.1. - Heavy Load Profile, No Control

Figure 4.2.1.2. - Heavy Load Profile, Baseline
46

Figure 4.2.1.3. - Heavy Load Profile, Load Sensing Control & Power Management

Figure 4.2.1.4 - Heavy Load Profile, Control with Daily Feedback
47

4.2.2. Medium Load User Profile
The Medium Load Profile has a lower measured peak load value and lower overall energy
use per person in comparison to the Heavy Load Profile. Conversely, it has a higher
measured peak load value and higher energy use per person than the Light Load Profile
when measured. Yet interestingly, the rated power is less than the Light Load Profile
rated power. Despite the differences, the output is not as high as the rated power and at
its highest, measured to be more than a third less. The parasitic load is less than the
Heavy Load Profile.
The output data for the Medium Load Profile can be seen in the figures below. They are
labeled according to the variables and shows the recorded data using average Watts (W).
In the following chapter, we will further analyze these load profiles by evaluating the
average daily load profile, which is calculated based on the average load in watts over
fifteen (15) minute intervals. This average load profile will also be used to compare
against typical commercial building load profiles to better understand the deviation
between the measured data and typical assumptions.
The data output during the "No Control" variant shows a constant demand requirement
that stays constant over the occupied and unoccupied demand, where the average watts
consumption is 65W/Person. The "Baseline" variant where the power management is
enabled shows a reduction but has the most significant with the "Daily Feedback" variant
where the calculated consumption is 24W/Person. Based on the output data, the overall
reduction achieved in the study case for the Medium Profile Load profile is 37%.
The power consumption during the unoccupied demand is the least with the
implementation of the APS, which is nearly twice the occupied hours. Due to the
48

building automation system set up that resets devices at 2am each day to allow for
systems updates to occur during unoccupied hours, it disables the function afterwards
for the equipment to completely shut down and results in continuous consumption of
power thereafter.
Due to the automatic building system setting that resets devices at 2am each day to allow
for systems updates to occur during unoccupied hours, it disables the function
afterwards to completely shut down, which in comparison to the Light and Ultra Light
Load Profile equipment types that typically is unplugged when not in use, will continue
to consume electricity thereafter. This is one of the main factors for this load profile to
have these higher loads, despite having lower rated power than the Light Load Profile.

49

Figure 4.2.2.1. - Medium Load Profile, No Control

Figure 4.2.2.2. - Medium Load Profile, Baseline
50

Figure 4.2.2.3. - Medium Load Profile, Occupancy Sensor & Power Management

Figure 4.2.2.4 - Medium Load Profile, Control with Daily Feedback
51

4.2.3. Light Load User Profile
The Light Load Profile has a lower measured peak load value and lower overall energy
use per person in comparison to the Medium Load Profile. Conversely, it has a higher
measured peak load value and higher energy user per person than the Ultra Light Load
Profile when measured. Yet interestingly, the rated power is higher than the Medium
Load Profile rated power. Despite the differences, the output is not as high as rated
power and at its highest, measured to be more than a third less. The parasitic load is less
than both the Heavy and Medium Load Profile.
The output data for the Light Load Profile can be seen in the figures below. They are
labeled according to the variables and shows the recorded data using average Watts (W).
In the following chapter, we will further analyze these load profiles by evaluating the
average daily load profile, which is calculated based on the average load in watts over
fifteen (15) minute intervals. This average load profile will also be used to compare
against typical commercial building load profiles to better understand the deviation
between the measured data and typical assumptions.
The data output during the "No Controls" variant shows less demand requirements
during the unoccupied demand in comparison to the High and Medium Load Profile,
where the average watts consumption is 28W/Person. The "Baseline" variant where the
equipment enables the power management controls shows a reduction but has the most
significant impact with the "APS and power management" variant where the calculated
consumption is 16 W/Person. . The "Daily Feedback" has a reduction compared to the
"Baseline" measure but shows a moderate increase compared to the "APS with power
52

management" due to the increase in its occupied demand. Based on the output data, the
overall reduction achieved in the study case for the Light Load Profile is 39%.
The power consumption during the unoccupied demand is the least with the
implementation of the APS, which is nearly twice the occupied hours. Despite the
building automation system set up that resets devices at 2am each day to allow for
systems updates to occur during unoccupied hours, the Light Load Profile is not affected
as the equipment is typically unplugged during this period. This is one of the main
factors for this user type profile to have these lower loads, coupled with having lower
rated power than the Heavy and Medium Load Profile.

53

Figure 4.2.3.1. - Light Load Profile, No Control

Figure 4.2.3.2. - Light Load Profile, Baseline
54

Figure 4.2.3.3 - Light Load Profile, Remote Switch Control & Power Management

Figure 4.2.3.4 - Light Load Profile, Control with Daily Feedback
55

4.2.4. Ultra Light Load User Profile
The Ultra Light Load Profile has the lowest measured peak value and lowest overall
energy use per person. The output is not as high as the rated power on the namplate
data. The parasitic load is lowest amongst the Heavy, Medium and Light Load Profiles.
The output data for the Ultra Light Load Profile can be seen in the figures below. They
are labeled according to the variables and shows the recorded data using average Watts
(W). In the following chapter, we will further analyze these load profiles by evaluating
the average daily load profile, which is calculated based on the average load in watts over
fifteen (15) minute intervals. This average load profile will also be used to compare
against typical commercial building load profiles to better understand the deviation
between the measured data and typical assumptions.
The data output during the "No Controls" variant shows less demand requirements
during the unoccupied demand in comparison to the Light Load Profile, where the
average watts consumption is 23W/Person. The "Baseline" variant where the equipment
enables the power management controls shows a reduction but has the most significant
impact with the "APS with power management" variant where the calculated
consumption is 11 W/Person and is near equivalent to the data output with the "Daily
Feedback" variant. Based on the output data, the overall reduction achieved in the study
design for the Ultra Light Load Profile is 10%.
The power consumption during the unoccupied demand is the least with the
implementation of the advanced power strips, which is nearly twice the occupied hours.
Despite the building automation system set up that resets devices at 2am each day to
allow for systems updates to occur during unoccupied hours, the Ultra Light Load Profile
56

is not affected as the equipment is typically unplugged during this period. This is one of
the main factors for this user type profile to have these lower loads, coupled with having
lower rated power than the Light Load Profile.

57

Figure 4.2.4.1. - Ultra Light Load Profile, No Control

Figure 4.2.4.2. - Ultra Light Load Profile, Baseline
58

Figure 4.2.4.3 – Ultra Light Load Profile, Timer Control & Power Management

Figure 4.2.4.4 – Ultra Light Load Profile, Control with Daily Feedback
59

4.3. PL Energy Use Summary
The results from the pilot study conducted for the four (4) load intensity classifications
with the various technology interventions identified that an energy use reduction of up to
39% can be achieved as compared to the baseline intervention.
The highest energy reduction percentage from the technology intervention was the
Remote Switch Control from the Light Load Profile with the result of 39% compared to
baseline. The results draws the conclusion that although laptop computers have less
power consumption than desktop computers, the monitor, if not properly shut down, can
be the factor to make up for the difference when not managed by the implementation of
advanced controls.
The highest reduction from the Daily Feedback intervention was the Occupancy Sensor
Control from the Medium Load Profile with the result of 37% compared to baseline. The
results draws the conclusion that although mini-tower desktops have less power
consumption than the standard towers, due to its compact equipment size can be placed
in difficult locations to access the on/off button, making it inconvenient for users to
reach to turn on unless daily feedback is provided to make users aware of its wasted
consumption.

60

4.3.1. PL Energy Use Summary for Reference
The table and the figures below summarizes the data output results from the pilot study
(see Figures 4.2.1.1. to 4.2.4.4.) and uses the benchmark evaluated in preceding chapters
(see Table 2.6) to highlight the results based on the average load (W) per Person over
time (h) and analyze the reduction rates by percentage (%) using the Baseline as its
comparative factor (see Figures 4.3.1.1. to 4.3.1.4). This is easier than reviewing each
data output of the set points that amounted to over 160,000+ for the analysis.

Average Watts/Person* (W)

Load User Profile
Type

Poor
(No Control)

Standard
(Baseline)

High
Performance
(Control/
Feedback)

Best
(Control/
Feedback)

Benchmark

190+

190

100

60

Heavy

176+

102

78

67

Medium

65+

38

37

34

Light

28+

27

22

16

Ultra Light

23+

13

12

11

* Load Factor is calculated based on values provided in Table 1.5. and is based on 40,000 ft² with 302 persons
Table 4.3.1.: Plug Load Factor based on W/Person

61

Figure 4.3.1.1 - Heavy Load Profile, Average Watt Hours/Person

Figure 4.3.1.2 - Medium Load Profile, Average Watt Hours/Person
62

Figure 4.3.1.3 - Light Load Profile, Average Watt Hours/Person

Figure 4.3.1.4 - Ultra Light Load Profile, Average Watt Hours/Person

63

4.4. Lessons Learned
This section identifies learning points from the various stages in the pilot study that
reflect on the experience to analyze and discuss what should be done in future activity to
avoid the pitfall or repeat the success.
4.4.1. People as Subject Matter
Having people as subject matters to conduct a study, requires extensive pre-planning.
The set up of monitoring devices shall be done in advance and preferably during periods
that the test subjects are not present. In addition, an identifiable research leader for the
study (especially in the case of computing equipment monitoring where personal devices
are involved) shall be available at all times during the test period as there are frequent
questions that arise from participants, regardless of the amount of education of the study
design or advanced notification is provided. Prior to the execution of the study,
preparation is recommended that includes a disclaimer to the test subjects that
cooperation is needed for a successful outcome as unexpected results may occur during
the test (i.e. unexpected shut down of equipment). Those conducting the study shall
expect that for every variable, there is a likelihood of an unexpected outcomes that was
not accounted for and thereby, it is critical that all characteristics of equipment and
controls are thoroughly examined prior to deployment. In addition, a detailed outline of
sequencing of events is required and with that, the test subjects shall be well informed of
each step of the process.
64

4.4.2. Understanding of Equipment Operability
The advanced technology discussed in this pilot study is not commonly implemented in
the commercial market. Thereby, having a thorough understanding of the
characteristics of the controls (see Section 3.3) prior to the start of the study is critical.
When the PL energy use monitoring is focused on desktop computers, it is highly
recommended that controls are coupled with the devices per Table 6.5. for the optimal
outcomes.
4.4.3. Unregulated = Not Highly Funded for Research
Due to PLs being unregulated, this pilot study was based on performance measurement
protocols that have not been developed previously. Everything from finding the
definition of associated terms to getting relevant information on standards and
guidelines came as a challenge, including searching for industry professionals that had
the extensive knowledge on the subject matter. The Appendix section (see Table A1) of
this thesis includes guidelines that were developed on performance measurement
protocols, thus allowing for this pilot study to be repeated for future activity.

65

Chapter 5 – Inferences from Data
This chapter analyzes the metered data taken from Chapter 4 to explore the following
topics: 1) compare the load profile developed from the measured data to the typical
eQuest default load profile, 2) calculate the percentage (%) deviation of PL factors for
and compares the results to the eQuest load profile, and 3) evaluates the results to
current methods of estimating PL energy use.
5.1. Modeled (Estimate) PLs vs. Measured (Actual) PL Energy
Use Overview
In Figure 5.1, two typical load profile data files from energy modeling software programs
for commercial buildings are presented and will be used as a baseline for the analysis
presented in this Chapter. These Load Profiles are used to determine the average Load
Factors (LF) or Part Load Factor (PLF), which are indicators of the variations in the
electrical load versus time. Often, engineers will use this information to assess electric
energy use by multiplying the peak load value with the part-load factor.
In this chapter, the eQuest Load Profile is used to compare against the actual measured
load profiles to analyze any deviations that occur while assessing the peak demand or
part load factor. Based on the detailed data evaluated in Chapter 4, evidence shows that
a substantial reduction can be seen in implementing advanced controls and daily
feedback in the Load Profiles to optimize energy use. In the following figures (See
figures 5.1.1.1 to 5.1.4.4), the measured results are further analyzed to study how the
implementation of these interventions currently compare against the computational
modeled environments using the typical Load Profile data file and is done so by
66

overlapping day-by-day data in 15 minute intervals (in comparison to the preceding
chapter which presented data in week-by-week in 1 minute intervals). The steps to arrive
to the results shown in these figures can be found in the Appendix section of this thesis
(see figures A1 to A16).

Figure 5.1 -Typical Computational Load Profiles for Commercial Buildings

67

5.1.1. Heavy Load User Profile
We will begin the review with the output data from the Heavy Load Profile with how it
compares to the computational modeled environment. The default eQuest Load Profile
data file has a Load Factor (LF) of 46% which will be considered as the baseline in this
chapter. In addition, all measured data will use its peak demand as its maximum LF
percentage (%) for the analysis.
The LFs of the measured Heavy Load Profile per the study design are as follows: 1) No
Control - 85%, 2) Baseline - 49%, 3) Control - 32%, and 4) Daily Feedback - 37%. As a
result, the following are the respective deviations from the baseline: 1) +39%, 2) +3%, 3)
-14%, and 4) -9%. Review Figures 5.1.1.1 to 5.1.1.4 and Table 5.2 for further information
on the results.
The deviation closest to the eQuest Load Profile is "Baseline" and furthest is "No
Control" in the Heavy Load Profile.
68

.
Figure 5.1.1.1 - Heavy Load Profile, No Control

Figure 5.1.1.2 - Heavy Load Profile, Baseline
69

Figure 5.1.1.3 - Heavy Load Profile, Load Sensing Control & Power Management

Figure 5.1.1.4 - Heavy Load Profile, Control with Daily Feedback
70

5.1.2. Medium Load User Profile
The next output data is from the Medium Load Profile with how it compares to the
computational modeled environment. The default eQuest Load Profile data file of has a
Load Factor (LF) of 46% which will be considered as the baseline in this chapter. In
addition, all measured data will use its peak demand as its maximum LF percentage (%)
for the analysis.
The LFs of the measured Medium Load Profile per the study design are as follows: 1) No
Control - 91%, 2) Baseline - 52%, 3) Control - 40%, and 4) Daily Feedback - 29%. As a
result, the following are the respective deviations from the baseline: 1) +45%, 2) +6%, 3)
-6%, and 4) -17%. Review Figures 5.1.2.1 to 5.1.2.4 and Table 5.2 for further
information on the results.
The deviation closest to the eQuest Load Profile is "Baseline" and "Occupancy Sensor
Control and Power Management” with the same deviation percentage (%) and furthest is
"No Control" in the Medium Load Profile.

71

Figure 5.1.2.1 - Medium Load Profile, No Control

Figure 5.1.2.2 - Medium Load Profile, Baseline
72

Figure 5.1.2.3 - Medium Load Profile, Occupancy Sensor Control & Power Management

Figure 5.1.2.4 - Medium Load Profile, Control with Daily Feedback
73

5.1.3. Light Load User Profile
The next output data is from the Light Load Profile with how it compares to the
computational modeled environment. The default eQuest Load Profile data file of has a
Load Factor (LF) of 46% which will be considered as the baseline in this chapter. In
addition, all measured data will use its peak demand as its maximum LF percentage (%)
for the analysis.
The LFs of the measured Light Load Profile per the study design are as follows: 1) No
Control - 19%, 2) Baseline - 17%, 3) Control - 11%, and 4) Daily Feedback - 14%. As a
result, the following are the respective deviations from the baseline: 1) -27%, 2) -29%, 3)
-35%, and 4) -32%. Review Figures 5.1.3.1 to 5.1.3.4 and Table 5.2 for further
information on the results.
The deviation closest to the eQuest Load Profile is "No Control” and furthest is "Remote
Switch Control & Power Management" in the Medium Load Profile.

74

Figure 5.1.3.1 - Light Load Profile, No Control

Figure 5.1.3.2 - Light Load Profile, Baseline
75

Figure 5.1.3.3 - Light Load Profile, Remote Switch Control & Power Management

Figure 5.1.3.4 - Light Load Profile, Control with Daily Feedback
76

5.1.4. Ultra Light Load User Profile
The last output data is from the Ultra Light Load Profile with how it compares to the
computational modeled environment. The default eQuest Load Profile data file of has a
Load Factor (LF) of 46% which will be considered as the baseline in this chapter. In
addition, all measured data will use its peak demand as its maximum LF percentage (%)
for the analysis.
The LFs of the measured Ultra Light Load Profile per the study design are as follows: 1)
No Control - 21%, 2) Baseline - 13%, 3) Control - 11%, and 4) Daily Feedback - 12%. As a
result, the following are the respective deviations from the baseline: 1) -25%, 2) -33%, 3)
-45%, and 4) -34%. Review Figures 5.1.4.1 to 5.1.4.4 and Table 5.2 for further
information on the results.
The deviation closest to the eQuest Load Profile is "No Control” and furthest is "Timer
Control & Power Management" in the Ultra Light Load Profile.

77

Figure 5.1.4.1 – Ultra Light Load Profile, No Control

Figure 5.1.4.2 – Ultra Light Load Profile, Baseline
78

Figure 5.1.4.3 – Ultra Light Load Profile, Timer Control & Power Management

Figure 5.1.4.4 – Ultra Light Load Profile, Control with Daily Feedback
79

5.2. Modeled vs. Measured Comparative Analysis
Below is a table that summarizes the deviation by percentage using the eQuest load
profile as the baseline case based on the results from the pilot study:
Rank Load Profile Variable Type Deviation (%)
1 (<10%) Heavy Baseline +3%
2 (<10%) Medium Baseline +6%
2 (<10%) Medium Control -6%
3 (<10%) Heavy Daily Feedback -9%
4 (<20%) Heavy Control -14%
5 (<20%) Medium Daily Feedback -17%
6 (>20%) Ultra Light No Control -25%
7 (>20%) Light No Control -27%
8 (>20%) Light Baseline -29%
9 (>20%) Light Daily Feedback -32%
10 (>20%) Ultra Light Baseline -33%
11 (>20%) Ultra Light Daily Feedback -34%
12 (>20%) Light Control -35%
13 (>20%) Ultra Light Control -35%
14 (>20%) Heavy No Control +39%
15 (>20%) Medium No Control +45%
Table 5.2 – Comparative Analysis of Modeled vs. Measured Deviation
80

5.3. Summary of Analysis

Figure 5.3. – Summary of Load Factor Deviation by percentage (%)

Based on the comparative analysis of the deviation of load factors, the results from the
pilot study indicates that the default load profile from eQuest closest reflects the load
profile for desktop computing equipment for Heavy and Medium load user profiles in
commercial buildings with the baseline variable.
The analysis also indicates that Light and Ultra Light load user profiles are not
accounted for in the default load profiles since all the outputs resulted in a deviation
greater than 25%. These findings leads to the discussion that more research is needed to
understand load factors for efficient equipment that integrates advanced controls and
feedback systems.
81

Figure 5.3.1. – Heavy Load Profile, Measured vs. Modeled Analysis

Figure 5.3.2. – Medium Load Profile, Measured vs. Modeled Analysis
82

Figure 5.3.3. – Light Load Profile, Measured vs. Modeled Analysis

Figure 5.3.4. – Ultra Light Load Profile, Measured vs. Modeled Analysis
83

5.4. Load Factor Analysis

Figure 5.4. – Load Factor Analysis (Watts per Person)

Based on the results from the average load (W) per Person over time (h) from the
preceding chapter (see Table 4.3.1.), load factors for all variables were calculated by
using the part-load factor evaluated (see Figures 5.1.1. to 5.1.4.4.) and people density of
30 ft²/person to compare against the commonly used 1.0 W/ft². Results from the
comparison indicates that plug loads modeled estimates were in fact less than the
measured data. This outcome reconfirms that estimating PL energy use utilizing
standardized load factors assumptions W/ft² with standardized load profiles result in
inaccuracies of its actual measured consumption. In order to better understand PL
energy use in buildings and more importantly to reduce the load, it is recommended that
84

targeting a goal based on average load (W) per Person over time (h) can help architects
and engineers to set a baseline that is measurable and allows to then devise a plan that
focus on PL reduction strategies to optimize energy efficiency.
5.5. Lessons Learned
This section identifies learning points from the various stages in the pilot study that
reflect on the experience to analyze and discuss what should be done in future activity to
avoid the pitfall or repeat the success.
5.5.1. Use Standard Load Profiles to Estimate Standard
Load Factors Only
The inferences from the data analyzed in this chapter show evidence that the standard
load profiles are designed to meet the actual measurements of standard desktop
computers with network default power management settings. High performance and
advanced controls should be excluded from the use of these standard load profiles as
results indicate a deviation of 35% for laptop computers with advanced control settings.
Further research and investigation is needed to estimate PL energy use with advanced
technology interventions.
5.5.2. Assumption of Standard Load Profiles
The standard load profile demonstrated in this chapter shows evidence that unoccupied
load consumption is estimated to be at its minimum. For standard commercial
85

buildings, this is unlikely the case where the unoccupied energy demands will be higher,
thus adding a layer of complexity to standardization of load profiles.

86

Chapter 6 - Miscellaneous Load Overview
This chapter explores the PL energy use for non-PC equipment that includes: 1)
appliances, 2) imaging equipment, and 3) audio visual equipment for a comprehensive
understanding of PLs in building. The goal of the exploration of these equipment use
were to further study the following topics: 1) typical usage patterns of equipment, 2)
comparative analysis between rated power from the nameplate data and peak demand
power, and 3) evaluate appropriate advanced controls for each equipment based on its
operability and applicability of systems.
It is important to study the remaining non-PC equipment classified in the miscellaneous
loads in commercial buildings to have a comprehensive understanding on the make-up
of the other significant percentage of PL energy use. The following figures (see Figures
6.1.1. to 6.3.1.) presented in this chapter are the results based on the average load (W)
over time (h) with a priority on energy consumption from Monday through Friday to
study equipment usage patterns based on one minute intervals.
6.1. Appliances
When considering appliances in commercial buildings, one of the key factors in reducing
PLs is to maximize space efficiency in shared areas. For the purposes of this thesis, the
equipment measured in the commercial building has break rooms with refrigerators,
microwaves, coffee makers, filtered water dispensers, ice maker and cappuccino makers
that each serve approximately 100 occupants.
87

6.1.1. Microwave

Figure 6.1.1 – Microwave, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the microwave is on
active mode for 3.95 hours with the remaining 116.05 hours consuming parasitic loads.
Since the equipment requires power consumption only as needed by the occupant, the
key strategy for reducing plug loads for the microwave is to eliminate the 3.2W to 3.4W
parasitic loads by the use of occupancy sensors.
88

6.1.2. Refrigerator

Figure 6.1.2 – Refrigerator, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the refrigerator is on
active mode 24/7 with two (2) cooling cycle periods that occur daily that fluctuate
between 54W to 60W over 18-20 minutes. Since this equipment requires continuous
power consumption, there are no controls that can shut-off the device, thereby an Energy
Star rated efficient equipment is the key factor in reducing plug loads for refrigerators.
89

6.1.3. Coffee Maker

Figure 6.1.3 - Coffee Maker, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the coffee maker is on
active mode for 50.03 hours and operates on five (5) to six (6) cycle periods daily that
fluctuate between 430W to 885W over 1 minute duration every 15-20 minutes. Since
this equipment requires power consumption during occupied hours on consistent basis
from 7:30am to 5:30pm daily, the key strategy for reducing plug loads for the coffee
maker is to eliminate the 1.1W to 1.5W parasitic loads and the power when not in use is
by the use of timer controls.
90

6.1.4. Ice Maker

Figure 6.1.4 - Ice Maker, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the ice maker is on
active mode for 36.4 hours and operates on four (4) to five (5) cycle periods daily that
fluctuate between 475W to 650W over 60 minute duration every 5 (five) hours with the
remaining period consuming no parasitic loads. Since this equipment requires
continuous power consumption, there are no controls that can shut-off the device,
thereby an Energy Star rated efficient equipment is the key factor in reducing plug loads
for the ice maker.
91

6.1.5. Filtered Water Dispenser

Figure 6.1.5 - Filtered Water Dispenser, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the hot/cold filtered
water dispenser is on active mode for 24/7 and operates on a cycle period daily that
fluctuate between 120W to 140W over 10 minute duration, 536W to 537W over 2 minute
duration with the remaining period consuming parasitic loads. Since this equipment
requires power consumption during occupied hours on consistent basis from 7:30am to
5:30pm daily, the key strategy for reducing plug loads for the filtered water dispenser is
to eliminate the 2.5W to 4W parasitic loads and power when not in use by the use of
timer controls.
92

6.1.6. Cappuccino Maker

Figure 6.1.6 - Cappuccino Maker, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the cappuccino maker is
on active mode for 0.5 hours and ranges in power consumption between 252W to
1,450W with no parasitic loads. Since this equipment is used minimally, the key
strategy for reducing plug loads for the cappuccino maker is to eliminate the equipment
all together.
93

6.2. Imaging Equipment
When considering imaging equipment in commercial buildings, one of the key factors in
reducing PLs is eliminate all personal use of equipment and specify all-in-one multiple
function devices to minimize quantity of equipment needed. For the purposes of this
thesis, the equipment measured in the commercial building has laser printers, high
volume multi-function laser copiers and inkjet oversize printers that each serve
approximately 15 occupants.

94

6.2.1. Laser Printers

Figure 6.2.1 - Shared Laser Printers, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the laser printer is on
active mode for 60 hours and operates between 15W to 62W on average with the
remaining period consuming 6.9W to 7W parasitic loads. Since this equipment requires
power consumption during occupied hours daily, the key strategy for reducing plug loads
for the laser printer is to eliminate the 6.9W to 7W parasitic loads by the use of timer
controls.
95

6.2.2. Laser Copiers

Figure 6.2.2 - Shared Multi-Function Laser Copier, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the laser copier is on
active mode for 24/7 and operates between 36W to 106W on average with the remaining
period consuming 71W continuously. Since this equipment requires power consumption
during occupied hours daily, the key strategy for reducing plug loads for the laser copier
is to eliminate the parasitic loads and power when not in use by the use of timer controls.

96

6.2.3. Inkjet Printers

Figure 6.2.3 - Shared Inkjet Oversize Printer, Average Watt Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the inkjet printer is on
active mode for 24/7 and operates between 47W to 95W on average with the remaining
period consuming 46W continuously. Since this equipment requires power consumption
during occupied hours daily, the key strategy for reducing plug loads for the inkjet
printer is to eliminate the parasitic loads and power when not in use by the use of timer
controls.

97

6.3. Audio Visual Equipment
When considering audio visual equipment in commercial buildings, one of the key
factors in reducing PLs is modify the power management settings and centralize the
locations of the equipment to maximize shared usage. For the purposes of this thesis,
the equipment measured in the commercial building has flat screen displays for the all
seven (7) conference rooms and in shared areas throughout the building.

98

6.3.1. 42" LED/LCD Flat Screen Display

Figure 6.3.1 – 42” LED/LCD Flat Screen Display with Computing Device, Average Watt
Hours

Of the Hour 1 through 24, Monday through Friday of 120 hours, the LED/LCD flat
screen display with Computing Device is on active mode for 24/7 and operates between
52W to 89W continuously. The key strategy for reducing plug loads for the LED/LCD
flat screen display is to eliminate the parasitic loads and power when not in use by the
use of occupancy sensors.

99

6.4. Miscellaneous Load Analysis
Below is a table that ranks highest to lowest of the Average PL Energy Use for the
miscellaneous office equipment (N=10) measured from the pilot study:

Figure 6.4. - Miscellaneous Load Summary
Rank Equipment Type Rated Power
(W)*
Peak Watts
(W)*
Average
Watts (W)
1 Coffee Maker 3,800W 1,634W 196.10W
2 Laser Copier 6,786W 1,233W 195.50W
3 Ice Maker 782W 803W 171.60W
4 Hot/Cold Filtered Water 800W 712W 106.4W
100

Dispenser
5 42" LED/LCD Flat Panel
w/ CPU
200W +240W 184W 87.74W
6 DeskJet Plotter 840W 233W 51.06W
7 Refrigerator 1,800W 616W 48.96W
8 Microwave 1,200W 1,739W 48.32W
9 Laser Printer 600W 1,233W 23.67W
10 Cappuccino Maker 1,720W 1,453W 5.96W
*For reference only
Table 6.4 – Rated Power vs. Peak Watts

6.5. Controls Evaluation
Based on the examination of the advanced technology intervention explored in the pilot
study, control recommendations have been made in Table 6.5 of the various equipment
tested. These recommendations are made based on the distinct characteristics and
parameters that each of the control types have with considerations of the equipment it is
coupled with. This table is intended to be used as a guide to avoid any unexpected shut
downs and/or outcomes that was experienced while conducting the pilot study.
101

Table6.5-Recommended Controls for Typical Office Equipment
102

6.6. Lessons Learned
This section identifies learning points from the various stages in the pilot study that
reflect on the experience to analyze and discuss what should be done in future activity to
avoid the pitfall or repeat the success.
6.6.1. Loads Most Appropriate for Controls
Based on the measured data, equipment classified in the miscellaneous loads from this
study are most appropriate to be on controls as they operate only during the hours of
operation with the exception of refrigerators and ice makers. With controls that have
automatic shut-off will eliminate the consumption of parasitic loads and operate to only
consume power when needed.
6.6.2. Rated (W) ≠ Peak (W)
Rated power or the nameplate power rating is often interpreted as the peak power output
for equipment but evidence from the research shows that in fact, it is not. For certain
equipment, the peak power (W) exceeds the nameplate (W) and vice versa with no
consistency or relationship to a typical diversity factor. It can also be stated that the
equipment manufacturer reported data is not consistent with identifying the difference
between rated, active, and standby power.
6.6.3. Specify Energy Star Rated Equipment
ENERGY STAR is a label that identifies products, such as office equipment to meet the
strict energy efficiency guidelines by the U.S. Environmental Protection Agency (EPA).
103

More than 50 types of products, including office equipment, lighting and appliances can
qualify for the ENERGY STAR rating to help reduce energy use. Because certain office
equipment use up to 24 hours a day, choosing ENERGY STAR qualified equipment for
especially the high rated equipment (Coffee Maker, Laser Copier and Icemaker) is
important as it consumes 20-50% less energy than standard equipment (EPA, 2008).

104

Chapter 7 – Conclusion
Plug Loads not only contribute to internal heat gain but also can be found in all building
types. As mentioned in the introduction of this thesis, as the dependency on the internet
for information and data transfer increases, the electricity demand will pose a challenge
to design and operate Net Zero Energy Buildings. This thesis investigates the reduction
of PL energy use through the implementation of advanced controls and studies
behavioral changes by providing daily feedback. Inferences from the collected data
formulates methods of estimating PL energy using Watts Per Person while evaluating the
findings to improve computational energy modeling.
The key research findings covered in this thesis can be highlighted into three (3) points:
1) Watt Hour (Wh)/Person Analysis - for every desktop computer replaced with a laptop
there is a reduction of up to 150Wh/Person and the lower the people density, the higher
the power density, 2) Comparative Analysis of Modeled vs. Measured Data - current
computational load profile standards are based on desktop computers without advanced
controls, and 3) Technology vs. Behavioral Intervention - implementation of APS
consistently resulted in reduction of PL energy use in comparison to daily feedback to
influence behavior.
By implementing daily feedback response systems, the behavioral component of this
thesis resulted in measurable reduction compared to the baseline case for all load
profiles. But based on the results, there is no prescriptive method of calculating
reduction measures for the behavior.
In conclusion, the discrepancy found between the measured (actual) and modeled
(estimate) data of PL energy use for the high performance commercial building
105

discussed in the earlier chapters can be attributed to the following factors: 1) computing
equipment not properly getting shut down when not in use, thereby consuming power
that was not accounted for, 2) miscellaneous equipment loads not properly accounted for
during unoccupied hours, 3) building operation hours not properly defined, 4)
unanticipated power surge that occur at 2am daily that ignites equipment from properly
shutting down, and 5) power management settings that are customized to perform less
than network default baseline settings.
7.1. Key Findings
In the table below are implications that have been made based on results from this pilot
study summarized as key finding for stakeholders involved in building energy and life
cycle assessment.
Stakeholder Key Findings

Architects/
Designers
- Use of “Watts Per Person” can help guide design teams to
understand plug load efficiency
Engineers /
Energy Modelers
- Typical eQuest office equipment loads and profiles are
applicable for standard computer use without the
implementation of any controls
- Rated power (or nameplate data) and peak power
demands in equipment are not necessarily similar values
and there is no real strong peak power ratio or
correlation
- Standard load profiles underestimates the energy
demand/use during unoccupied hours as findings
indicate that many computer equipment and appliances,
imaging and audio visual equipment are not shut down
- Manufacturer provided equipment cut sheets have no
106

consistency on reporting of power consumption to
estimate peak
Building Owners /
Tenants
- Implementing a power management setting on
computing equipment to reduce energy can save up to
83% of power consumption
- Implementing APS on computing equipment to reduce
energy can save up to 37% of power consumption
- Replacing desktop computer with laptops can reduce PL
energy consumption of up to 150W/person
Utility Companies - For incentives/rebates offerings, expand plug load
controls beyond occupancy sensor types and include
other advanced controls that result in higher percent
energy use reduction
Researchers - To be consistent with how energy use is assessed with
utility companies, use average 15 minute intervals when
metering devices when measuring and verifying data
- Refer to performance guidelines provided in the
Appendix for recommended protocols when measuring
PLs for buildings
All - Not all APS are appropriate for all devices (assess the
operability of each equipment carefully and refer to
Table 6.4. for additional guidance)
Table 7.1 – Pilot Study Implications Summary
7.2. Standard Load Profiles Re-Examined
The findings from this pilot study indicate that the typical eQuest Load Profiles used for
energy modeling software to estimate plug loads in commercial buildings are designed
for standard desktop computers without the implementation of any controls or laptop
devices. Based on the pilot study, the results with the highest PL reduction load profiles
were extracted to re-examine those load profiles for reference.
107

Figure 7.2.1. – Heavy Load Profile, Load Sensing Control & Power Management

Figure 7.2.2. – Medium Load Profile, Control with Daily Feedback
108

Figure 7.2.3. – Light Load Profile, Remote Switch Control & Power Management

Figure 7.2.4. – Net Zero Load Profile, Timer Control & Power Management
109

Chapter 8 – Future Work
This pilot study had its limitations due to the number of meters available, but with
funding that may become available in the future, outlined below are additional research
points that should be further investigated.
8.1. Analysis of Other Building Types
Examining the plug load use breakdowns for other project types including homes, retail,
schools, data centers, and hospitality and healthcare facilities will help to make this
thesis more comprehensive. Since plug loads are building user specific, further research
for each will be needed to understand reduction measures from its base case.
8.2. Increase Sampling per Building Type
Due to the limitation of meters and controls available for this research during the three
(3) month test period allotted in the study design, future work will be to extend the test
duration to allow for at minimum one (1) month logging per variable with a 5-10%
sampling size per load intensity classification per building type. It will also be to
measure periodically after the one (1) month reporting to verify whether reduction rates
remain constant after the study design has been conducted.
8.3. Evaluate Load Profiles per Advanced Controls
For the purposes of this study, multiple APS were used to evaluate its efficiency and
effectiveness. To further the evaluation, each load profile shall be tested on all the APS
110

for a comprehensive analysis on any variables that may occur based on the different
advance controls.
8.4. Advanced Controls for Miscellaneous Loads
Based on the recommended controls identified in Chapter 6, further testing on the
equipment classified in the miscellaneous loads should be investigated to determine %
reduction using controls.
8.5. Incentives as a Factor for Further Reduction
Presently, utility rebate programs are offered through utility companies such as Southern
California Edison (SCE) that offers $15/sensor plug load occupancy rebates for non-
residential customers, regardless of size and energy use for plug load reduction through
these advanced control implementations. The option of having these incentives as an
intervention variable was not included in this thesis. With that, a study shall be
conducted to measure if having incentives (i.e. monetary reward) could further reduce
plug load energy use in buildings.
8.6. Plug Loads as a Priority in Building Design
The priority of reduction strategies for PL energy use is relatively low when designing
buildings highly due to its complexity. To achieve maximum energy use savings in
buildings will require design teams to evaluate PL energy use with considerations of
implementing advanced control strategies. Rather than estimating the PLs using
conventional methods based on default plug load factors of W/ft², design teams are
111

encouraged to use the methodology of estimating PL use by the average load of
W/Person.
112

Bibliography:
ASHRAE 90.1, 2010. Energy Standard for Buildings Except Low-Rise Residential
Buildings I-P Edition, American Society of Heating, Refrigerating and Air-
Conditioning Engineers Inc.
ASHRAE Handbook, 2009. Fundamentals I-P Edition. American Society of Heating,
Refrigerating and Air-Conditioning Engineers Inc.
California Energy Commission, 2008. 2008 Building Energy Efficiency Standards for
Residential and Nonresidential Buildings. CEC.
Catherine Mercier and Laura Moorefield, 2011. Commercial Office Plug Load Savings
and Assessment: Executive Summary, Ecova (Mercier , Moorefield, 2011)
Chad Lobato, Shanti Pless, Michael Sheppy and Paul Torcellini. 2011. Reducing Plug and
Process Loads for Large Scale, Low Energy Office Building: NREL’s Research
Support Facilty. Paper presented at the ASHRAE Winter Conference in Las
Vegas, NV, January 29-February 2, 2011 (Lobato, Pless, Sheppy, Torcellini, 2011)

Charles Eley, John Arent, Deborah Stanescu and Kristen Salinas, 2009. Rethinking
Percent Savings. Prepared by Codes and Standards Development, Building
Programs Unit and Architectural Energy Corporation (Eley, Arent, Stanescu,
Salinas, 2009)
CPUC, 2008. California Long Term Energy Efficiency Strategic Plan; Achieving
Maximum Energy Savings in California for 2009 and Beyond. California Public
Utilities Commission.
City of Zurich, 2011. On the Way to the 2,000-Watt Society: Zurich's Path to
Sustainable Energy Use. Office for Environmental and Health Protection Zurich
COMNET, 2010. Commercial Buildings Energy Modeling Guidelines and Procedures,
RESNET. Commercial Energy Services Network.
David Kaneda, Brad Jacobson and Peter Rumsey. 2010. Plug Load Reduction: The Next
Big Hurdle for Net Zero Energy Building Design. Paper presented at ACEEE
Summer Study on Energy Efficiency in Buildings (Kaneda, Jacobson, Rumsey
2010)
ENERGY STAR, 2007. Energy Savings Summary of ENERGY STAR Computer
Specifications. ENERGY STAR. Retrieved from http://www.energystar.gov/
Hootman, Tom, 2013. Net Zero Energy Design: A Guide for Commercial Architecture.
Ian Metzger, Dylan Cutler and Michael Sheppy, 2012. Plug Load Control and
Behavioral Change Research in GSA Office Buildings (Metzger, Cutler, Sheppy,
2012)
113

Jeff Haberl, Hywel Davies, Brendan Owens and Bruce Hunn, 2008. ASHRAE's New
Performance Measurement Protocols for Commercial Buildings. Proceedings of
the Eighth International Conference for Enhanced Building Operations, Berlin,
Germany (Haberl, Davies, Owns, Hunn, 2008)
Michael Murray, 2012. Plug Loads in Commercial Buildings. Paper presented at the
ASHRAE winter conference (Murray, 2011).
S. Mark Fisher, Nabeel Sultan and Ryan C. Stromquist, 2006. Plug Load Reduction for
a Net Zero Energy Building. Paper presented at ACEEE Summer Study on
Energy Efficiency in Buildings (Fisher, Sultan, Stromquist 2006)
Spencer Sator, 2008. Managing Office Plug Loads. E Source Energy Managers'
Quarterly Newsletter for Second Quarter II f June 2008 (Sator, 2008).
Tom Hootman, David Okada, Shanti Pless, Michael Sheppy and Paul Torcellini, 2012.
Net Zero Blue Print. Publication in High Performance Buildings, Fall 2012.
(Hootman, Okada, Pless, Sheppy, Torcellini, 2012)
U.S. Green Building Council, 2009. LEED Reference Guide for Green Interior Design
and Construction, 2009 Edition. U.S. Green Building Council.
U.S. Energy Information Administration, 2008. Electrical Power Annual 2008. U.S.
Energy Information Administration. Retrieved from http://www.eia.gov/
U.S. Environmental Protection Agency, 2010. Energy Star and Other Climate
Protection Partnerships: 2010 Annual Report. U.S. EPA.
Voss, Karsten and Eike Musall. 2011. Net Zero Energy Buildings. German National
Library (Detail Green Books 2011)
Yamda Zhang, Charlotte Bonneville, Neha Arora and Randall Higa. Integrated Lighting
and Plug Load Controls. Paper presented at ACEEE Summer Study on Energy
Efficiency in Buildings (Zhang, Bonneville, Arora, Higa 2012)

114

Appendix

Pilot Credit: Plug Load Reduction

Applicable Rating Systems
This credit is available for pilot testing by the following LEED project types:
 New Construction
 Core and Shell
 Schools
 Retail: New Construction
 Retail: Commercial Interiors
 Commercial Interiors
 Existing Buildings: Operations + Maintenance

Intent
To achieve increasing levels of energy conservation by establishing performance
measurement criteria to reduce excessive plug load energy use.

Requirements
Meet the applicable recommendations and standards for plug loads in Chapter 4, Design
Strategies and Recommendations in the ASHRAE 50% Advanced Energy Design Guide
AND
115

Reduce the connected plug load energy use below the baseline building performance by
implementing at least six (6) of the following strategies, categorized into two sections: 1)
Technology and 2) Design. At least two (2) from the Design category must be met.
The items listed below apply to computers, imaging equipment, audio visual and
appliances (unless otherwise noted). Excluded are HVAC, lighting, and building
envelope products.
Technology
 Use ENERGY STAR rated laptop computers for 90% of computer devices*
 Use USB LED-backlit LCD displays for 90% of computer monitors*
 Use soft phone technology with DigitalLine VoiP for >60% of the telephones
 Use LED task lights with lamps that last at least 24,000 hours on occupancy
sensors for 100% of the task lights
 Use Advanced Power Strips (APS) controls for 90% of imaging equipment, audio
visual and appliances*
 Use Advanced Power Strips (APS) controls for 90% of computers*
 Consolidate 100% of individual imaging equipment (copiers, scanners, printers,
fax machines) to All-In-One machines
 Set the network power management control to power savings mode for 90% of
computers at minimum when plugged in to: 2 minutes to dim the display (for
laptops only), 5 minutes to turn off the display and 10 minutes to put the
computer to sleep*
Design
 Integrate Advanced Power Strips (APS) for 90% of the private office desks,
systems furniture and conferencing tables to enable single switch shutdown or
automatic shut off for the connected plug loads*
 Design to consolidate break rooms to be centralized on projects to service one
per floor by eliminating all kitchenettes and coffee areas
 Design at minimum ratio of 60 occupants for all shared imaging equipment by
eliminating all individual printers, copiers, fax machines
 Design at minimum 60% of occupants in regularly occupied spaces to have
unassigned seating to promote mobility by dedicating spaces as follows:
116

- Roaming (>50% of time at Desk - Unassigned), >60% of occupants
- Anchor (<50% of time at Desk – Assigned), <40% of occupants
- Mobile (2-3 Days Away from Office – Unassigned), Included in Roaming %
- Remote (>3 Days Away from Office - Unassigned), Included in Roaming %
*Strategies indicated with an (*) are required to use the performance measurement protocol provided in the following
section as part of the documentation
Performance
Demonstrate the reduction of plug loads through measurement and verification methods
by following the performance measurement protocol provided below.
The collected data must demonstrate performance results that is better than the baseline
measurements provided below for each of the load intensity classifications specified on
the project.
Load Intensity Classification
- Heavy (100Wh/Person): Desktop Computer with Monitor(s)
- Medium (40Wh/Person): Mini Tower Desktop Computer with Monitor
- Light (30Wh/Person): Laptop Computer with Monitor
- Ultra Light (15Wh/Person): Laptop Computer Only OR Laptop Computer with
USB LED-backlit LCD monitor

Step 1 – Establish Load Intensity Classification(s)
Conduct an equipment audit and categorize user type profiles based on the load intensity
classifications provided above
Step 2 – Set up Metering Devices to Perform Measurements
Using plug load meters that can collect data, measure 5% (minimum) of the occupants
from each load intensity classification category that is applicable to the project for a
minimum of two (2) continuous weeks (exclude weeks with Holidays) at fifteen (15)
minute intervals, using continuous Watts (W) output data from the meter to calculate
the W/Person by averaging the measured load (W) over the total period of time (h)
(excluding weekends)
Step 3 – Evaluate the Measured Data
117

Using the collected data, calculate using the following method:
Extract the data for the 24 hour period from Monday through Friday (12am to 12 am) for
the weeks measured and average the energy use of the equipment to calculate the Watt
Hours (Wh) as described in Step 2
Step 4 – Required Documentation
For compliance, provide the following documents:
 Complete office equipment list for the project with quantity, make/model,
manufacturer specifications with name plate data
 Raw data output file from the metering device supplemented with the average
energy use calculation spreadsheet
 Graph output from the metering device with measured dates and intervals
 Building hours of operation

Credit Submittals
General:
1. Register for Pilot Credit(s) here.
2. Register a username at LEEDUser.com, and participate in online forum
3. Submit feedback survey; supply PDF of your survey/confirmation of
completion with credit documentation

118

Credit Specific:
1. For All strategies in Technology, provide an equipment schedule that
includes quantity, make/model and nameplate data with manufacturer's
product information and cut sheets
2. For all strategies in Design, provide floor plans highlighting relevant
information and provide a narrative with the methodology of how
strategies were pursued
3. For all Advanced Power Strips (APS), provide plans and schedules
denoting control types by equipment, locations with specifications
4. For strategies denoted with (*), refer to performance guidelines provided
above for required documentation
Additional Questions
1. Do the criteria associated with plug loads align with your projects and
occupant user needs?
2. Please identify why your project opted not to address the strategies that
were not pursued from hardware/design?
3. How difficult was it to document this credit?

Background Information
This pilot credit is proposed to expand on the current equipment and appliances,
controllability of lighting requirements, in order to address the further reduction of plug
loads

Table A1 – Plug Load Reduction, LEED Pilot Credit
119

Appendix-Data

Figure A1 - Heavy Load Profile, No Control - Average Watt Hours

Figure A2 - Heavy Load Profile, Baseline - Average Watt Hours
120

Figure A3 - Heavy Load Profile, Control - Average Watt Hours

Figure A4 - Heavy Load Profile, Daily Feedback - Average Watt Hours
121

Figure A5 - Medium Load Profile, No Control - Average Watt Hours

Figure A6 - Medium Load Profile, Baseline - Average Watt Hours

122

Figure A7 - Medium Load Profile, Control - Average Watt Hours

Figure A8 - Medium Load Profile, Daily Feedback - Average Watt Hours
123

Figure A9 - Light Load Profile, No Control - Average Watt Hours

Figure A10 - Light Load Profile, Baseline - Average Watt Hours
124

Figure A11 - Light Load Profile, Control - Average Watt Hours

Figure A12 - Light Load Profile, Daily Feedback - Average Watt Hours
125

Figure A13 - Ultra Light Load Profile, No Control - Average Watt Hours

Figure A14 - Ultra Light Load Profile, Baseline - Average Watt Hours
126

Figure A15 - Ultra Light Load Profile, Control - Average Watt Hours

Figure A16 - Ultra Light Load Profile, Daily Feedback - Average Watt Hours

Abstract (if available)

Abstract As building envelopes have improved due to more restrictive energy codes, internal loads have increased largely due to the proliferation of computers, electronics, appliances, imaging and audio visual equipment that continues to grow in commercial buildings. As the dependency on the internet for information and data transfer increases, the electricity demand will pose a challenge to design and operate Net Zero Energy Buildings (NZEBs). Plug Loads (PLs) as a proportion of the building load has become the largest non-regulated building energy load and represents the third highest electricity end-use in California's commercial office buildings, accounting for 23% of the total building electricity consumption (Ecova 2011,2). In the Annual Energy Outlook 2008 (AEO2008), prepared by the Energy Information Administration (EIA) that presents long-term projections of energy supply and demand through 2030 states that office equipment and personal computers are the ""fastest growing electrical end uses"" in the commercial sector. ❧ This thesis entitled “Watts Per Person"" Paradigm to Design Net Zero Energy Buildings, measures the implementation of advanced controls and behavioral interventions to study the reduction of PL energy use in the commercial sector. By integrating real world data extracted from an energy efficient commercial building of its energy use, the results produce a new methodology on estimating PL energy use by calculating based on ""Watts Per Person"" and analyzes computational simulation methods to design NZEBs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Zero net energy institutional building

PDF

Net zero energy building: the integration of design strategies and PVs for zero-energy consumption

PDF

Impact of occupants in building performance: extracting information from building data

PDF

Zero peak homes: designing for zero electric peak demand in new single family residential buildings sited in California climate zone 10

PDF

District energy systems: Studying building types at an urban scale to understand building energy consumption and waste energy generation

PDF

Net-zero cultural urbanism: the implementation of traditional and cultural net-zero urban housing design in Lagos

PDF

Landscape and building solar loads: development of a computer-based tool to aid in the design of landscape to reduce solar gain and energy consumption in low-rise residential buildings

Asset Metadata

Creator Kim, Mika Yagi (author)

Core Title "Watts per person" paradigm to design net zero energy buildings: examining technology interventions and integrating occupant feedback to reduce plug loads in a commercial building

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Publication Date 06/28/2013

Defense Date 06/28/2013

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

Tag advanced controls,Architecture,ASHRAE,Building Science,buildings,Commercial,LEED,miscellaneous loads,net zero energy,OAI-PMH Harvest,occupant feedback,paradigm,parasitic loads,plug loads,Watts

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Konis, Kyle (committee chair), Hall, Ken (committee member), Schiler, Marc (committee member)

Creator Email mikayagi.leedap@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-281325

Unique identifier UC11294526

Identifier etd-KimMikaYag-1717.pdf (filename),usctheses-c3-281325 (legacy record id)

Legacy Identifier etd-KimMikaYag-1717.pdf

Dmrecord 281325

Document Type Thesis

Format application/pdf (imt)

Rights Kim, Mika Yagi

Type texts

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

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

Repository Name University of Southern California Digital Library

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