The Role of Modeling When Designing For Absolute Energy Use Intensity Requirements in A Design-Build Framework
The Role of Modeling When Designing For Absolute Energy Use Intensity Requirements in A Design-Build Framework
The Role of Modeling When Designing For Absolute Energy Use Intensity Requirements in A Design-Build Framework
Conference Paper
NREL/CP-5500-49067
March 2011
The submitted manuscript has been offered by an employee of the Alliance for Sustainable Energy, LLC
(Alliance), a contractor of the US Government under Contract No. DE-AC36-08GO28308. Accordingly, the US
Government and Alliance retain a nonexclusive royalty-free license to publish or reproduce the published form of
this contribution, or allow others to do so, for US Government purposes.
This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof, nor any of their employees, makes any warranty,
express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of
any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately
owned rights. Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation,
or favoring by the United States government or any agency thereof. The views and opinions of authors
expressed herein do not necessarily state or reflect those of the United States government or any agency thereof.
P.O. Box 62
Oak Ridge, TN 37831-0062
phone: 865.576.8401
fax: 865.576.5728
email: mailto:[email protected]
Cover Photos: (left to right) PIX 16416, PIX 17423, PIX 16560, PIX 17613, PIX 17436, PIX 17721
Printed on paper containing at least 50% wastepaper, including 10% post consumer waste.
The Role of Modeling When Designing for
Absolute Energy Use Intensity
Requirements in a Design-Build Framework
ABSTRACT
The Research Support Facility at the National Renewable Energy Laboratory is a 220,000-ft2 office building designed to
serve 822 occupants, to use 35.1 kBtu/(ft2yr), to use half the energy of an equivalent minimally code-compliant building, and
eventually to produce as much renewable energy annually as it consumes. These goals and their substantiation through
simulation were explicitly included in the fixed price design-build contract. The energy model had to be repeatedly updated
to match design documents and the final building, as it was built, to the greatest degree practical. Computer modeling played
a key role in diagnosing the energy impacts of program and decisions and in verifying that the contractual energy goals
would be met within the specified budget. The primary tool used was a whole-building energy simulation program. Other
simulation tools were used to provide more detail or to complement the primary tool as required by the delivery schedule,
including tools to calculate thermal bridging, daylighting, natural ventilation, data center energy consumption, transpired
solar collectors, thermal storage in the crawlspace, and electricity generation by photovoltaic panels. Results were either fed
back into the main whole-building energy simulation tool or used to post-process model output to provide the most accurate
annual simulations possible. This paper details the models used in the design process and how they informed important
program and design decisions from design to completion.
INTRODUCTION
The National Renewable Energy Laboratorys (NREL) mission is to advance the U.S. Department of Energys and the
nations goals in the areas of energy security, environmental quality, and economic vitality. From the beginning, NREL
recognized that its new Research Support Facility (RSF) represented a unique opportunity to demonstrate the state-of-the-art
in energy-efficient, cost-effective commercial office design, construction, and operation. Today, buildings use roughly 40%
of total primary U.S. energy consumption (22% residential, 18% non-residential); energy consumption in this sector is
projected to grow by 25% in the next two decades 1. The RSF is intended to prove that significant gains in energy efficiency
can be realized cost effectively in commercial buildings with available technologies if careful attention is paid to project
energy goals, the delivery process, and integrated design. We examine the details of how computer simulation tools were
1
http://buildingsdatabook.eren.doe.gov/docs/xls_pdf/1.1.3.pdf
Adam Hirsch, Shanti Pless, Rob Guglielmetti, and Paul Torcellini are engineers in the commercial building group at the National
Renewable Energy Laboratory, Golden, CO. David Okada and Porus Antia are engineers at Stantec Engineering, San Francisco, CA.
1
used to help design the RSF and what capabilities the project required of those tools. We also present a portrait of how setting
an absolute whole-building energy consumption target changes the role of energy modeling. Several teams were involved in
the competition phase; however, we focus on the activities of the design-build team that was awarded the contract.
PROJECT OVERVIEW
The RSF project contains several novel features related to the delivery model, team structure, and Request for Proposals
(RFP). Descriptions of how energy modeling was used in the design process follow.
Delivery Model
NREL decided early that to deliver the RSF, with its challenging performance requirements, on time and on budget, a
traditional design-bid-build procurement process would not suffice. Rather, a performance-based Best Value Design-
Build/Fixed Price with Award Fee delivery approach (Post 2010) was pursued to encourage innovation, reduce risk,
expedite construction and delivery, control costs, make optimal use of team members expertise, and establish measurable
success criteria. The RSF procurement strategy provides an important context for understanding the design process and the
use of energy modeling tools. By hiring a design-build team, NREL encouraged an integrated design process comprising
architects, engineers, and builders working toward well-defined goals. This arrangement resulted in an iterative design
pattern involving the entire team. Detailed computer simulations were used to assess whether the design as it evolved would
meet performance requirements. It also required NREL to clearly define the scope and goals in the RFP and then allow the
design-build team to find creative solutions. Having specific end-use and whole-building energy use goals necessitated that
energy modeling be included in the design process from the beginning. The fixed budget for all work (conceptual design,
preliminary design, final design, and construction) of $64 million, formulated by the U.S. Department of Energy, was
determined before energy goals were established. This fixed budget, coupled with the energy goals, required that cost
modeling be emphasized as much as energy modeling. The selected design-build team performed a great deal of conceptual
phase energy modeling in the months leading to the submission of its design competition proposal in March 2008.
Preliminary design occurred from July 2008, when the contract was awarded, through November 2008. Final design took
place from January 2009 through July 2009. Construction began in February 2009 and lasted until June 2010, when it was
turned over and ready for occupancy.
Objectives
In the conceptual documents of the original RFP (dated February 6, 2008), the objectives were prioritized and then
divided into three groups: Mission Critical, Highly Desirable, and If Possible. Competing design-build teams were judged
based on their ability to meet as many objectives as possible while meeting the overall budget constraint. A subset of those
goals related directly to energy modeling and the low energy design process includes:
1. Mission Critical
a. LEED Platinum designation
b. ENERGY STAR appliances, unless another system outperforms
2. Highly Desirable
a. 800 staff capacity
b. 35.1 kBtu/(ft2yr) 2
c. Measurable 50% energy savings versus ASHRAE 90.1-2004
3. If Possible
2
Absolute energy goals of 25, 32, and 35.1 kBtu/(ft2yr) all appear at certain points in the project documents; 25 kBtu/(ft2yr) assumes 650 occupants and
data center energy use prorated to reflect that only a portion of the data center services are consumed in the RSF; 32 kBtu/(ft2yr) assumes 822 occupants
and prorated data center energy; 35.1 kBtu/(ft2yr) includes the entire data center energy consumption.
2
a. Net zero energy building approach
b. LEED Platinum Plus
c. Exceed 50% savings over ASHRAE 90.1-2004 baseline
The absolute site energy consumption, net zero energy balance, and LEED Platinum goals in particular influenced the
modeling tools and design process. The RFP also specified parameters such as temperature and humidity set points, nighttime
setback, maximum U-value of windows, and compliance with elements of ASHRAE 55 that have energy impacts.
3
http://apps1.eere.energy.gov/buildings/publications/pdfs/commercial_initiative/energy_use_intensity_targets.pdf
3
into the main whole-building energy model or used to post-process its output to provide an integrated picture of whole-
building energy use. For example:
Daylighting calculations were performed with detailed lighting simulation tools, including window shading and light
redirection; the daylight performance and correlated electric lighting dimming and switching responses were fed into
the whole-building energy model;
Thermal effects of natural ventilation strategies were analyzed using a separate whole-building energy modeling tool
with more advanced thermal analysis capabilities and used to post-process the results of the main energy analysis
model;
Data center energy use, including electricity consumed by information technologies equipment, cooling energy, fan
energy, and heat recovery, was modeled outside of the main energy model and added to its energy projections;
A separate model of air heating by transpired solar collectors on the south face of the building was used to post-
process the energy modeling results;
Ventilation preheating benefits of the crawlspace during the heating season, with inputs from the data center and
transpired solar collector models, were modeled using a separate finite-difference thermal model and combined with
the energy model output;
Renewable energy generation by PV panels was modeled separately.
4
Daylighting/Lighting Modeling
A lighting simulation tool was used to quantify annual daylighting characteristics and to establish optimal photosensor
placement for energy savings. It is driven by a physically based, backward ray-tracing simulation designed to provide
accurate quantitative and qualitative daylight and electric lighting predictions, even when considering complex fenestration
systems such as daylight redirection devices. The lighting tool was initially employed in the design competition phase in the
context of the LEED daylight credit iEQ8.1, mandated in the RFP. Analysis of the workplane illuminance under clear skies at
noon on the equinox a metric for assessing compliance with this criteria showed that even with best practices of
separating view glass and daylight glass and using a daylight redirection device to reflect incoming sunlight deeper into the
building, a maximum floor depth of 60 could be daylit sufficiently to satisfy the requirements. This consideration effectively
set the buildings footprint. For the energy modeling, the lighting analysis tool was able to provide an 8,760 hour schedule of
the lighting power fraction for an electric lighting system with dimming and switching controls to match the specified office
illuminance set point. This schedule can be passed seamlessly to the whole-building energy model. For this analysis, the
typical office space was divided into a south perimeter zone, a core zone, and a north perimeter zone. Other daylit spaces
were simulated using the built-in continuous dimming sensor and daylighting calculations available in the energy model.
Installed lighting power density in open office areas is very low only 0.62 W/ft2, facilitated by the open office structure (no
obstructions), highly reflective room surface finishes, ability to use a regular grid, and efficient electrical lights.
5
occupied hours and be used during occupied hours to preheat ventilation air. A no-economizer case was also modeled
employing a runaround loop to transfer heat from the data center hot aisle to preheat domestic hot water and ventilation air.
More efficient data center equipment would reduce the amount of heat available for ventilation preheating; however, future
efficiency gains were assumed to be matched by increased data center us,e so the overall load would remain the same.
6
developed to modify lighting schedules for typical RSF office spaces, only 4 could be added before the energy model code
reached its maximum length. As a workaround, lighting files representing south perimeter and core spaces in both wings
were used to model all those spaces, and the energy models built-in daylighting controls were used for the north perimeter
spaces in both wings. Layout and fixture changes in the electrical lighting design were incorporated into the energy model as
they occurred. For the final round of energy modeling, seven lighting files were incorporated into the model as other
schedules were combined.
Natural ventilation modeling analysis of the fitness center, connector wing, and information commons (library) was also
performed in the core and shell design development, implementing the lessons learned in the preliminary modeling. The
effectiveness of night purging was analyzed for different window arrangements to identify the best configuration of manual
and automatic windows and their control for effective cooling. We found that natural ventilation would likely be ineffective
in the connector wing, mainly because of high transient internal loads (for example, densely occupied conference rooms)
relative to the amount of perimeter faade available for ventilation. Also, the full exterior building geometry (rather than that
of a typical wing) was eventually included in the natural ventilation model to more realistically capture shading. The
technique for quantifying the cooling savings for natural ventilation was changed from adjusting the thermostat set point on
select days to post-processing cooling energy results on select days. The design team also tried to capture the influence of
occupant behavior vis-a-vis window operation by halving the potential energy savings from natural ventilation to reflect the
possibility that it would be suboptimal.
Thermal Bridges
In the same way small details have a large impact on simulated energy use, small architectural details can affect the
energy use bottom line by compromising the envelopes thermal properties. Two-dimensional heat transfer analysis and fin
heat transfer formulas were used to study thermal bridging in the balconies at the east and west ends of the RSF. A detailed 2-
D heat transfer model was used to derate the U-values of the constructions used in the energy model for the whole-building
energy model, to investigate design strategies to minimize the bridging, and identify likely areas of problematic
condensation. Areas initially studied included the connection of the superstructure to the substructure and balcony attachment
points. In all cases the design-build team adopted mitigating measures. Later this analysis was used to evaluate the insulation
required between the radiant ceiling slabs and the adjacent supply air floor plenums and prove the necessity of avoiding a
specific standard detail for mounting photovoltaics (PV) to the roof. This drove a complete change of the roofing system to a
standing seam strategy to avoid penetrations. The standing seam structure was affixed to the roof membrane; the funds came
from the project contingency.
Baseline Simulation
The project RFP included an absolute energy use intensity target for energy consumption, a net zero energy balance
target, and a minimum 50% energy savings versus a minimally code-compliant equivalent building with the same
unregulated loads and program. This requirement necessitated construction of a baseline building model, built according to
the requirements laid out in Appendix G of ASHRAE 90.1-2004, which include distributing glazing in equal bands around
7
the building, eliminating shading devices, modeling a standard HVAC system type with airside economizers and screw
chillers with a coefficient of performance of 4.9, and using standard lighting power densities per space type. U-values of the
opaque envelope and glazing assemblies were taken from ASHRAE 90.1-2004 Table 5.5-5. In the baseline case, computer
workstations were assumed to consume 120 Watts versus the 65 Watts in the low energy case.
Closeout Modeling
As construction neared completion, the energy model was again updated to reflect minor modifications and new
information available from the construction. Changes include pump and fan power as balanced, revised assumptions for plug
load equipment, and in some cases measured parasitic power draws of specific components.
LESSONS LEARNED
One of the most important lessons of the RSF modeling effort is the degree to which unregulated plug and process loads
can dominate energy use, and the degree to which these loads must be understood and controlled to meet a stringent absolute
whole-building energy use target. Energy modeling typically focuses on envelope properties, mechanical systems, and
regulated internal loads, but for the RSF great attention was paid to details such as schedules and the energy use of individual
items ranging from telephones and task lights to lighting control systems (a constant 728 Watts in the case of the latter),
distribution transformer losses, white noise generators, and sump pumps. Architectural details such as thermal bridging in
window mullions (which might typically be ignored) were also important considerations. When detailed calculations of
window performance were done instead of taking overall U-values from ASHRAE Fundamentals (Chapter 15 table 4),
expected U-values were found to be 30% greater.
With a whole-building absolute energy use goal, plug and process loads cannot be assumed to be the same between
baseline and low-energy models and therefore fall out of the analysis when the two models are compared. They must be
carefully modeled before construction and measured when the building is occupied. A corollary to this lesson is that the
building owner, who controls the building program, must be committed to the energy goals and monitor end use energy
consumption over time to make building performance match design and to sustain those energy savings. A design team
cannot deliver a building that automatically reaches aggressive whole-building energy use targets; the building must emerge
from an integrated design team that places energy use intensity at the top of the checklist throughout and uses rigorous
simulation to verify the designs performance at every milestone.
The energy modeling process required close coordination between the engineering and design teams because of the
expedited design-build process and the RFPs strict substantiation requirements. There was a constant tension between the
model detail required for accuracy and the compressed design-build schedule. In retrospect, the energy modeling process
should have been included as significant constraint instead of being squeezed into a schedule that was based primarily on
design documentation and construction. This would likely have resulted in fewer design changes, a more streamlined
decision making process, and a more integrated building. The workaround calculations, such as the heat budget of the
labyrinth, were critical to the energy modeling process, but were significantly more time consuming than originally
anticipated. When approaching these types of calculations in the future, the team would be much more likely to model these
in an existing energy model with detailed heat transfer simulation capabilities rather than completely from scratch. Modeling
uncertainties led the team to build an energy-use contingency into its design to provide a cushion should some assumptions
fail. The RSF closeout energy modeling report cites a final energy use of 33.3 kBtu/(ft2yr), 5.9% better than NRELs goal of
35.1 kBtu/(ft2yr).
Energy modeling played a dramatically different role in this project versus a project where energy modeling is required
only for code compliance or voluntary certification programs. The team had to model early and often at a high level of detail,
to gain confidence that the building as constructed would meet its energy use goals. Energy modeling fundamentally
influenced the design of the building, including the narrow floor plate; the size and performance of windows; architectural
detailing to avoid thermal bridging; the data center cooling approach; and office equipment.
8
CONCLUSION
Energy modeling helped the RSF design-build team to assess from the outset whether the design would be able to
accomplish NRELs ambitious energy efficiency goals. The modeling process pushed the teams available energy modeling
toolkit to, and perhaps past, the limit of its capabilities. Separate models were required for daylighting, natural ventilation,
data center energy use, crawlspace heat recovery and storage, and renewable energy generation. The team successfully knit
these outputs together; ultimately, however, a tool that could simulate the RSF from start to finish more seamlessly and with
a reasonable learning curve would have been advantageous. Detailed monitoring of the RSF operational energy use has
begun and the building is now partially occupied. These data will be used to verify model assumptions, calibrate a more
detailed EnergyPlus model, and quantify operational energy use across days, seasons, and years.
REFERENCES
ASHRAE, 2008. 2008 ASHRAE Enironmental Guidelines for Datacom Equipment Expanding the Recommended
Environmental Envelope. 11 pp. American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.,
Atlanta,GA. http://tc99.ashraetcs.org/documents/ASHRAE_Extended_Environmental_Envelope_Final_Aug_1_2008.pdf
Griffith, B., Torcellini, P., Long, N., Crawley, D., and J. Ryan. 2006. Assessment of the Technical Potential for Achieving
Zero-Energy Commercial Buildings: Preprint. 15 pp. NREL Report No. CP-550-39830.
Lobato, C., Pless, S., Sheppy, M., and P. Torcellini. 2010. Reducing Plug and Process Loads for a Large Scale Ultra Efficient
Office Building: NRELs Research Support Facility. 8 pp. NREL Report No. CP-550-49002.
Post, N. 2010. Green Design-Build Model Crafted for Buildings To Achieve Net-Zero Energy Use.
http://enr.ecnext.com/coms2/article_busu100505NetZeroEnerg-1
Torcellini, P., Pless, S., Deru, M., Griffith, B., Long, N., and R. Judkoff. 2006. Lessons Learned from Case Studies of Six
High-Performance Buildings. 151 pp. NREL Report No. TP-550-37542.
Torcellini, P., Pless, S., Lobato, C., and T. Hootman. 2010. Main Street Zero Energy Buildings: The Zero Energy Method in
Concept and Practice: Preprint. 12 pp. NREL Report No. CP-550-47870.
9
Form Approved
REPORT DOCUMENTATION PAGE OMB No. 0704-0188
The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this
collection of information, including suggestions for reducing the burden, to Department of Defense, Executive Services and Communications Directorate (0704-0188). Respondents
should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a
currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ORGANIZATION.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
March 2011 Conference Paper
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Role of Computer Simulation in Designing an Energy Efficient DE-AC36-08GO28308
Building: Preprint 5b. GRANT NUMBER
11. SPONSORING/MONITORING
AGENCY REPORT NUMBER
16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
OF ABSTRACT OF PAGES
a. REPORT b. ABSTRACT c. THIS PAGE
Unclassified Unclassified Unclassified UL
19b. TELEPHONE NUMBER (Include area code)
F1147-E(10/2008)