We know that we need to turn off the lights when we leave the room, but what about the heating and air conditioning? Lee Soon JaeAn assistant professor in the Department of Civil and Mineral Engineering at the University of Toronto’s School of Applied Science and Engineering, he discovered that artificial intelligence (AI) could bring better progress.
Lee’s latest research project, Grid Interactive Smart Campus Building, is a three-year project aimed at reducing the University of T’s climate footprint by leveraging AI to optimize heating and cooling systems in existing buildings. is a pilot project. This project will be implemented in cooperation with: Chigun LeeProfessor of Mineral Engineering, jointly funded by the University of T’s Climate Positive Energy and Climate Positive Campus initiatives.
“Buildings account for about 25 to 30 percent of energy consumption and energy sector greenhouse gas emissions in Canada and globally,” Lee says.
“Given that people spend an average of 90% of their lives indoors, ensuring a comfortable and healthy indoor environment is an important function of building systems. We should be able to take advantage of it.”
Lee’s research addresses this problem by applying AI solutions to architectural science. In the first year of the project, the team will focus on creating a digital twin (virtual representation) of the test center at 255 McCall Street.
In the next phase, researchers will develop deep reinforcement learning algorithms for optimal control of heating and cooling systems. The algorithm is pre-trained on the digital twin to avoid over-stressing the real building.
The algorithm is then implemented in a real test center and further fine-tuned through interaction with the building. If successful, Lee hopes to use the same approach to transform more campus buildings into smart buildings and contribute to T University’s low-carbon action plan.
“Sixty percent of the energy consumption on the St. George campus comes from heating and cooling the buildings,” he says.
Lee’s research group is also investigating how humans interact with buildings in a project funded by NSERC Discovery on Scalable Systems for Intelligent and Interactive Buildings. It’s an emerging research area with relatively few published studies, and Lee wants to change that.
While previous techniques have relied on data such as the correlation between the thermostat’s set temperature and other parameters (such as the time of day), Lee and his team instead focus on causal relationships, such as the occupant’s intentions regarding the thermostat. We take advantage of the factors that influence our decisions. Set Temperature – Develop reliable, human-centric smart solutions.
“If we can understand how humans interact with buildings in a causal context, we can make buildings more intelligent and human-interactive,” Lee says.
Lee is not the only researcher interested in using machine learning and AI techniques in buildings, but the field is open to others, such as the automotive and healthcare industries, as individual buildings have vastly different energy consumption profiles and needs. lagging behind other sectors. .
“A customized solution for one building may not translate perfectly to another building,” says Lee.
“This is a major obstacle to making buildings smarter. If AI can be seamlessly combined with existing building science discipline knowledge, it will be scalable and reliable to create sustainable buildings. We can build better solutions.”
To tackle this problem, the team is partnering with PLC Group and funded by the Ontario Innovation Center to develop scalable digital twinning tools for building energy systems. If this tool is effective, it will provide the building industry with a solution for creating intelligent, interactive and more sustainable buildings around the world.
“The use of AI in building management systems not only has the potential to significantly improve the sustainability of our built environment, but also the way we interact with it,” Lee said. says Mr.
