Optimizing energy use with digital twins

Heating and cooling are some of the largest sources of building energy use, from family homes to large multi-residential buildings to critical infrastructure such as hospitals. In 2019, carbon emissions from building operations made up 21 per cent of the year’s total global greenhouse gas emissions, according to estimates by the International Energy Agency (IEA). The IEA is an intergovernmental organization that provides energy sector policy recommendations, data and analysis to support energy security and a transition to clean energy.

Professor Jenn McArthur, a Toronto Metropolitan University (TMU) building science researcher and mechanical engineer in the Department of Architectural Science, is developing and applying digital twin technologies to optimize building performance, including the reduction of energy consumption. Digital twins are virtual versions of physical objects or systems connected to live building data to reflect physical reality in real-time. This digital infrastructure allows users to conduct scenario testing, assessing potential improvements and solutions before implementation through data analytics and mathematical modelling. 

Energy savings results 

At a B.C. hospital in the summer of 2024, professor McArthur created a digital twin to explore how to decrease cooling-related energy consumption. Together with industry collaborator H.H. Angus and Associates, they tested the impact of changing the settings of the hospital’s cooling tower and chiller. The cooling tower sends water to the chiller, which ejects the heat. Using the digital twin to model scenarios, they developed a plan to run the cooling towers more aggressively so the chiller, which is the bigger energy consumer, could reduce its workload, resulting in overall energy savings. 

“It’s been tested, and we have improved the chiller performance by 33 per cent, saving approximately 22 per cent of the total cooling energy,” said professor McArthur. “We haven’t physically touched anything. All we’ve done was change the controls based on what we were able to test in the digital twin.” As part of the summer pilot, the hospital ran the updated controls for six hours a day. At the end of September, the revised settings became the hospital’s new normal, in place 24 hours a day. The team will look to see if similar energy savings are possible at some Ontario-based hospital sites in the summer of 2025. 

Professor McArthur is also building on previous work in boiler optimization to reduce heating energy consumption. This work was done in collaboration with Parity, a software services company that focuses on HVAC, and together they were able to find substantial energy savings – as high as 40 or 50 per cent – in large, multi-residential buildings. She is looking to apply that boiler knowledge to the hospital setting during the cooler weather. 

Other energy-savings-related digital twins work includes ongoing work with TMU facilities. Professor McArthur has been working on a digital twin platform of the TMU campus with the university’s facilities management team. She hopes to apply her research as part of the university’s decarbonization working group. 

Read “The benefit of noise-injection for dynamic gray-box model creation” in Advanced Engineering Informatics on enhancing modelling accuracy in heating and cooling system energy optimization.  (external link, opens in new window) 

This digital infrastructure allows users to conduct scenario testing, assessing potential improvements and solutions before implementation through data analytics and mathematical modelling.

The research described in this article is supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and H.H. Angus and Associates. The Smart Campus Integrated Platform was supported by NSERC and FuseForward Solution Group, and earlier work noted in the article was supported by NSERC and Parity Inc.