Studio: A “Model Approach” to Sustainable Design
October 01, 2005
Appeared in Environmental Design + Construction
In the design competition for the new $65-million, 232,000-square-foot U.S. Environmental Protection Agency (EPA) Region 8 Headquarters in downtown Denver, competitors knew only the location and size of the site, area of the building in square feet, and the client’s primary goal: that the design meet the U.S. Green Building Council’s criteria for Leadership in Energy and Environmental Design (LEED) silver certification within the constraints of the project budget and schedule. Working with the architects, the Zimmer Gunsul Frasca Partnership and Opus Northwest LLC, Los Angeles-based Syska Hennessy Group performed energy simulations to determine the optimal site orientation and massing of the building to reduce energy consumption and capture daylight for quality of life benefits, as well as the architectural aesthetic of the building envelope.
The project team began with a computer simulation of the weather at the site, developing a map of the solar path and prevailing wind direction and speed. Then they built a computer model of the building to assess the relevant energy-efficiency measures — including high-performance glazing and building envelope construction, external solar shading devices, and daylight-responsive lighting controls — and estimate the number of LEED® points for optimized energy performance.
As a result of this integrated effort, the team won the job.
This is an example of the power of energy simulation — when used as part of an integrated design process — to enable a design team to achieve high-performance, sustainable design.
Energy simulation is not new; experienced practitioners have used computer simulation tools for at least 20 years. Complex energy simulations evolved from simple HVAC peak load calculations used by mechanical engineers to size heating and air conditioning equipment. Today’s sophisticated tools not only can be used to size equipment, but also to simulate the performance of building systems throughout the year and facilitate decision-making based on life cycle cost analyses.
The early code-oriented programs, such as DOE-2, a freeware building energy analysis program developed by the Department of Energy (DOE) to calculate energy use and cost for all types of buildings, had a less than user-friendly front end. Private contractors have developed more user-friendly, Windows-based energy simulation interfaces, such as eQUEST (jointly developed by Hirsch and the U.S. DOE) and VisualDOE (developed by Eley Associates). These tools are designed to make it easy for a user to input data — whether that means plugging in numbers, grabbing information from drop-down menus, or using an interface to import drawings from an AutoCAD file. They serve as the front end for newer versions of DOE’s calculation engine, DOE-2.1e or DOE-2.2.
The modeling process begins with a comparison of the energy efficiency parameters of the “as-designed” building with a baseline case that complies with the minimum standards set by local or state code, such as California’s Title 24, or the ASHRAE 90 standard. The model typically compares the effects of parameters in three basic categories — building envelope, lighting and HVAC — on annual energy consumption (Mbtu/yr), annual CO2 emissions (lbs/yr) and annual energy cost ($/yr). (Note that the U.S. Green Building Council bases its LEED rating system on energy cost, not energy consumption.)
Building envelope and lighting efficiency measures are typically modeled first. For the building envelope, for example, the engineer might compare dual-pane clear glass and dual-pane Low-E glass. Lighting simulations might compare lower lighting power density (i.e. 10-percent reduction) and daylight-responsive controls. Daylighting, which is affected by the glazing system, shading system and lighting control system, is a shared metric involving the building envelope and electrical lighting.
Based on the results of the building envelope and lighting models, a cumulative model of the building envelope and lighting measures are run to form a new baseline (Baseline 2). Baseline 2 becomes the baseline for comparison of HVAC measures, such as increased efficiency equipment and alternative HVAC systems such as under floor air distribution or ground coupled cooling and heating. Then a cumulative model is run of all the HVAC metrics and compared with Baseline 2. This yields the final “design case.”
For example, Syska Hennessy Group analyzed the International Center for Possibility Thinking at the Crystal Cathedral Ministries Campus, Garden Grove, Calif. Designed by Richard Meier & Partners, New York City, the four-story, 50,000-square foot, hospitality center has a skin of clear glass and stainless steel and a 65-foot-high atrium with an integrated HVAC/natural ventilation system. Energy efficiency measures (EEMs) for building envelope, lighting and HVAC were compared with a minimally compliant Title 24 base case. The energy simulation process yielded a design case that reduced energy consumption, annual emissions and annual energy costs, each by 31 percent when compared with the Title 24 base case.
These software tools can be used to run simplified energy models early in the schematic phase to help with building orientation, massing, and so forth. But to gain the full benefits of energy simulation in designing high performance buildings, the process should be integrated into the design process and updated continuously as information is developed throughout project design — from schematic design to construction documents. Moreover, the process yields the greatest benefits when it is used to measure the performance of the building as a whole, rather than measuring individual components such as the lighting or chillers or motors.
It should be noted that data input must be accurate in order to yield reliable results — the “garbage in, garbage out” (“GIGO”) rule applies to energy simulation as it does to every other domain of computing. Also, every computer program has limitations, and the user must understand these, or else it is easy to misinterpret the results. Finally, it should be remembered that the energy simulation tools are no substitute for sustainable, high performance design expertise – they merely supplement it.
There are many additional advanced tools available for designers of high performance
buildings - Radiance, a public-domain lighting and daylighting software developed
by Lawrence Berkley Labs, and Air-Pak, a thermal analysis and computational
fluid dynamics (CFD) software developed by Fluent, are two examples of many
that are now available to the design community. CFD modeling typically is used
in the design of critical facilities such as operating rooms, patient isolation
rooms, and data centers, where airflow patterns are critical to maintaining
a sterile environment or one which meets narrow tolerances for temperature
and humidity, and for large volume spaces such as atria and arenas.
In the case of the Crystal Cathedral Hospitality Center’s complex atrium space, CFD modeling was used to evaluate air flow patterns, temperature distribution and thermal comfort in two modes: air conditioning during a peak summer day and natural ventilation (no wind) on a spring or autumn day. The results were used to fine-tune and optimize the performance of the natural ventilation and mechanical systems – including decisions about location of air diffusers, sizes and locations of natural ventilation inlet and outlet openings, control strategies, and so forth.
Until recently, these advanced tools had a steep learning curve. For example, Radiance was original designed to operate in a Unix computing system environment; however, most engineers use a Windows environment. That meant installing a virtual Unix box on the PC and learning seashell scripting to run the model. To make the process easier, programs such as Ecotect have integrated a module of Radiance, enabling the user to control the application within an interface rather than going through scripting. This limits the number of parameters to some extent, but it allows testing of the relative performance of daylighting systems. It is a practical way to obtain the key advantages of Radiance.
Some in the design community are now looking at simulation packages such as Virtual Environment by Integrated Environmental Solutions Ltd., which ties a lot of these programs together. The engineer develops a single front end as an input module (i.e., the base building), and then is able to export that model to other modules. For example, the model could be exported to Energy Plus (developed by The Department of Energy) to calculate energy, or to Radiance to calculate daylighting levels. The developers of the Virtual Environment package have also linked the front end user input with a building information system model, enabling the user to input capital cost data on equipment and materials, and use it to calculate and compare capital and lifecycle costs. Most designers have to use separate front end models for performing HVAC load calculations, energy simulations, daylight models, CFD models, etc., so having tools with a single front end model is a significant development for the time-constrained design community!
As these kinds of developments reduce the time required to input data and enhance the power of the results, advanced building performance simulation will become not only desirable but also feasible for every project — not only those pursuing LEED certification. When used as part of an integrated design process, this “model approach” supports the development of high-performance, sustainable design for every project. This benefits owners by making every building more cost-effective to operate and maintain, while providing design guidance used for enhancing the quality of life for a building’s occupants. And it benefits the global community by reducing consumption of finite natural resources.
Rob Bolin, P.E., LEED AP, is senior vice president, Sustainable Design, and Arvinder Dang, LEED AP, is sustainable design specialist for Syska Hennessy Group, Los Angeles.