Should Energy Professionals Rely on Energy Modeling for Efficiency Projections?

Energy modeling can be a helpful method of projecting how much power a building or household uses based on specific parameters. Yet, as the world works toward achieving its sustainability goals, it raises the question of whether it is reliable enough.

The following explores energy modeling’s advantages and limitations for ensuring accurate predictions, including user behaviors and passive design. Professionals will also understand the importance of real-world datasets and continuous monitoring for the most reliable outcomes.

Caution: Applying Laboratory Results to Larger Populations

Most energy professionals have likely seen impressive results from technologies promising better efficiency. However, 2022 research regarding smart thermostats indicated promising lab results do not necessarily equate to what one can expect on a larger scale. 

The study’s authors examined two experiments involving people from 1,385 households who either received a two-way programmable smart thermostat or kept their conventional models. The researchers then analyzed participants’ energy consumption for 18 months, compiling more than 16 million hourly electricity use records and nearly 700,000 daily natural gas consumption observations. 

After reviewing the results, the team discovered smart thermostats may increase gas consumption by 4.2% and electricity usage by 2.3%. Researchers concluded they don’t statistically or economically impact energy consumption in a significant way. 

Energy modeling demonstrated better results for smart thermostat users but failed to account for interactions with these products. For example, when researchers examined almost 4 million users’ climate control systems and the associated interactions, they found people frequently overrode energy-efficiency settings, including permanently scheduled temperature settings. 

Most smart thermostat users also canceled the available energy savings benefits. People who want to see the most advantages must strategically learn about and use the gadget’s features for optimal savings — even if it means making lifestyle changes. 

The study’s results are similar to utility bill calculations, which vary based on personal preferences and needs. Although Americans pay an average of $489.35 in monthly utility costs, the totals vary by location and resource consumption. 

Reducing the Energy Burdens of Low-Income Households

How much could passive design elements increase energy-efficiency gains for low-income households? Researchers at the University of Notre Dame knew people in this group already dealt with significant energy burdens. They wanted to understand how passive design choices could help these individuals but lacked sufficient data.

The team used artificial intelligence to predict a household’s energy expenses based on their home’s passive design, achieving an accuracy rate of over 74.2% in forecasting power burdens. The researchers also used demographic data, including 1,441 census tracts, to build their model.

This study resulted in three passive design choices with significant impacts on a home’s energy efficiency — the percentage of shading the building receives, the size of the windows, and whether the windows were an openable or fixed style. Choosing well-insulated windows to maximize heat flow resistance could also improve results. 

Part of the research involved using a convolutional neural network and analyzing Google Street View images. This method was easy to scale and much more efficient than previous energy audit methods requiring one-at-a-time building assessments in a given area. 

Although the researchers know these conclusions will not solve poverty, they believe their work supports actionable improvements. They also plan to investigate the impacts of additional passive energy solutions, such as cool roofs, green roofs and insulation.

This energy modeling effort saw positive results, and the enormous dataset is likely one of the reasons why. The more reliable information professionals have, the better their chances of getting trustworthy outcomes for real-life situations. 

Broadening the Data Available for Energy Modeling 

Data limitations make energy modeling efforts less accurate. For example, if professionals lack information about a customer’s state or do not have models specific to particular buildings, even the most careful attempts are off-base and unhelpful to those who ordinarily use them for improved decision-making. 

However, increasing the amount of high-quality data can lead to more reliable modeling outcomes. To decarbonize the nation’s buildings, researchers from the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL) created a new dataset to show how buildings use — or could potentially use — energy. 

Before this effort, the options to make estimates were limited and costly. Plus, data limitations restricted attempts to study aggregate building stock loads or see diversified loads’ effects on models, forcing analysts to resort to simplified evaluations. 

This new dataset enhanced baseline load profiles, comparing them to electric load data from 11 utilities and 2.3 million power meters. The NREL researchers also had state-based information about natural gas usage and supplemental data from the U.S. Energy Information Administration to inform their work. 

Data collection also included hundreds of thousands of energy modeling specifics showing how buildings currently use resources and what could change with efficiency and electrification upgrades. The datasets also update every six months, making the content useful as people assess how to implement energy-efficient upgrades. 

Approximately 40 people worked on this project for four years, making it a considerable undertaking. Fortunately, the individuals working on it were part of a national lab and had significant expertise. 

Is Energy Modeling Worthwhile?

These examples highlight why professionals cannot hastily assume energy modeling is worth — or not worth — pursuing. They must evaluate the specifics on a case-by-case basis, considering factors such as the amount, quality and recentness of the available data. 

Considering the data’s source is also necessary. Did it come from a manufacturer’s small-scale test or researchers not affiliated with the specific product promising better energy efficiency?

Energy industry workers must also discuss how usage habits affect a home’s energy efficiency. Teaching customers easy but effective ways to get the most out of their chosen upgrades is an excellent option for helping them establish habits to maximize results. 

The efficiency projections from any given model are only as good as the data used to create it. The overall accuracy is usually higher when the information closely reflects the real world and the people hoping to gain from efficiency-related changes.