Hi Every Energy Model Is Wrong—And Here Is Why They Are Indispensable.
Recently, LEED has come under fire for accounts of certified buildings not performing as well as their energy models predicted. Frequently mentioned amongst the antagonistic “gotcha” coverage is an out-of-context 2007 quote by the USGBC Research Committee acknowledging: “Buildings have a poor track record of performing as predicted during design.”
Within context, the research committee clarifies the reasons for the frequency of underperforming energy models, citing “inaccurate or improperly used analysis tools, lack of integration of complex interconnected systems, value engineering after design, poor construction practices, no building commissioning, and incomplete or improper understanding of operations and maintenance practices.” Not nearly an exhaustive list, but all legitimate considerations.
Energy models will continue to become more accurate as the market develops and methodologies and software become more robust and sophisticated. Models can be calibrated based on actual performance data to further increase their accuracy for measurement and verification purposes. But let’s be clear—to some degree, all energy models are wrong. They always will be. At first blush, one may reasonably presuppose energy models are based solely on physics and, as such, they should be extremely precise—perhaps 95 to 99 percent accurate. Yet all building energy models also require inputs based on assumptions and long-term trends. We cannot predict the future—e.g., abnormal weather patterns, mechanical malfunctions, changes in occupancy, occupant behavior—but all of these factors have a chaotic effect on performance outcomes.
Nevertheless, energy modeling is essential for any high-performance building project—no matter how big or small. Energy models facilitate sustainable design in three essential ways:
1. To Understand. Energy models allow us to understand more about how our buildings are likely to perform. They allow design teams to test hypotheses and simulate field conditions for both proposed designs and existing structures. I was once approached to advise on dripping water in the ceiling of a museum. Through energy modeling—specifically a hygrothermal (i.e., pertaining to both humidity and temperature) analysis—it was determined that an ill-advised vapor retarder was preventing vapor drive toward the exterior. Add seasonal temperature extremes and high interior relative humidity, and it was a recipe for condensation.
2. To Compare. It is easier (and much less expensive) to experiment in the computer than in real life. Energy models are most valuable during the earlier stages of the design process when their results can help guide decision-making. As a parametric design tool, energy models can be used to evaluate everything from conceptual massing options to different glass types. This is the very premise of the “simple box” energy modeling analysis within the LEED v4 integrative process credit—and there is an abundance of user-friendly software platforms currently available in the market, many of them free. This kind of early-stage design performance modeling allows design teams to go beyond rules of thumb to actually fine-tune environmental control systems and energy conservation measures.
3. To Forecast. Buildings are investments, and the separation between construction capital and operating expenses makes it difficult to finance long-term improvements in building performance. Energy models improve our insight of the connections between—and business-case benefits of—various building systems in relation to high-performance outcomes. Despite a certain degree of imprecision, energy models can be leveraged to forecast the return on investment in high-performance building upgrades, such as onsite renewable energy, automated exterior louver systems or even that extra inch of rigid insulation on the roof. More frequently, project teams are using energy models to anticipate the order of magnitude to which future climate change could impact the economics of building performance, operations and maintenance.
In a recent TED talk, climate modeler Gavin A. Schmidt, director of the NASA Goddard Institute for Space Studies, insisted, “Models are not right or wrong; they’re always wrong. They’re always approximations. The question you have to ask is whether a model tells you more information than you would have had otherwise.”
Energy models are not meant to predict the future. They are powerful tools that enable us to better understand the behavior of our structures, fine-tune building systems and strategies, and forecast future performance trends.
.jpg)
