For most of us, office cooling and heating systems are like wallpaper: you only really notice them if they catch fire.

For Omid Ardakanian, office buildings offer potential energy savings on a skyscraper scale.

“My obsession, if you want to call it that, comes from the sensors in these buildings,” said Ardakanian, a computing science professor in the Faculty of Science. “If you have sensors, you have data.”

Omid Ardakanian

Omid Ardakanian

And if you have data, the world is your oyster.

Modern commercial buildings contain thousands of sensors that measure everything from the plumbing to the temperature in different rooms to whether a specific door is open or closed or the lights are on.

This information is collected by the building management system (BMS), which can be programmed to do things like turn on the air conditioning when a building gets hot.

But as anyone who has ever frozen in their office cubicle in July can attest, these systems are far from foolproof.

Enter Ardakanian and his research team.

They’ve developed software that uses machine learning to make office buildings more comfortable while cutting greenhouse gas emissions and lowering operating costs.

The software uses data from a building’s BMS to make ‘smart’ heating, cooling and lighting systems smarter. Less energy is wasted heating or cooling buildings past the point of comfort or when they’re unoccupied.


Artificial intelligence will kill jobs – and create them by Steven Globerman


Ardakanian said the software works best when it knows the location of every occupant in a building but that raises privacy concerns.

“Because most people-counting solutions use intrusive techniques like detecting and tracking people with RGB cameras, the data collected by these sensors can be potentially used to make sensitive inferences, such as who is where at any given time.”

Ardakanian’s research team is looking at refining his software to address those concerns – a major step toward moving it into widespread use.

One method they’re considering is using non-intrusive techniques such as analyzing data from carbon dioxide sensors or thermal arrays in buildings to get accurate estimates of how many people are in the building and when. They’re also studying how machine learning might be used to determine the number of occupants without compromising privacy.

Tests and simulations his research team has run suggest energy consumption could be cut by at least 10 per cent in most buildings – and in some cases, much more. That’s a significant boost to the balance sheet over a building’s lifetime.

But the benefit goes even further. About 40 per cent of all the energy an office building uses goes to heating and cooling to keep occupants comfortable.

“Ten per cent is a lot if you’re able to do it at scale,” said Ardakanian. “If we could do that for the millions of office buildings in North America, there’s a potential for significant impact in terms of energy use reduction and occupant comfort.”

| By Lewis Kelly

This article was submitted by the University of Alberta’s online publication Folio, a Troy Media content provider partner.

© Troy Media


heating, emissions

The views, opinions and positions expressed by columnists and contributors are the author’s alone. They do not inherently or expressly reflect the views, opinions and/or positions of our publication.