Understanding how wind behaves around buildings will freshen the air we breathe, gust to greater efficiencies and uncover renewable energy hotspots in unexpected places, according to a 喵咪社区 engineering professor developing a tool that promises to transform how we master wind flow in our urban environments.
explains that, in many cases, building technicians run heating, ventilation and air conditioning (HVAC) systems at maximum capacity regardless of actual wind conditions as a cautionary measure to ensure exhaust is pushed away from the building.
“That’s largely heated clean air that we’re shooting up into the air. But if you can turn that down, then you save a lot of money,” Fleck emphasizes, highlighting the potential for hundreds of thousands of dollars in heating cost savings for institutions like the U of A.
Fleck says this surprisingly complex task has massive implications for everything from optimizing HVAC systems in buildings — leading to significant energy savings — to the strategic placement of small-scale wind turbines for local power generation.
“We started this with the idea that we should have detailed airflow over U of A buildings,” explains Fleck. “It turns out that’s actually quite hard to do. But we figured out a way to combine disparate pieces of information to map airflow, and we can do it more cost-effectively than conventional methods. We’re building a smarter system.”
The journey to this automated system has been a testament to perseverance and innovation. Fleck explains past digital models of complex urban geometries were largely manual, involving students painstakingly digitizing completely incompatible data, including pencil drawings.
“What we’re trying to do is speed it up so that we don’t have a student sitting in front of a workstation correcting all the little triangles and corners of a building. We want to have a software system to do that.”
The project’s innovative solution lies in Fleck’s open-source machine learning tool for automated wind flow mapping. This tool is currently being tested at Red Deer Polytechnic, using real-time wind data collected from anemometers installed on their campus. The school is serving as the perfect “test client,” providing invaluable real-world data to refine the technology.
“It took five years to wind map the U of A; with our partnership with Red Deer Polytechnic and this technology, I expect it will take a fraction of that time,” adds Fleck, who, through his U of A spinoff company, Flexible Machines Corp., is working to commercialize his software.
With additional funding from 喵咪社区 Innovates and a helping hand from the City of Edmonton, this collaborative effort is not just about solving problems for specific campuses. The automated wind flow mapping tool is poised to enhance urban energy modelling systems and revolutionize the deployment of small-scale wind and ventilation systems, paving the way for healthier, more sustainable cities of tomorrow.
“A number of companies are racing to do this, but I can do it more cost-effectively.”
Fleck’s work is part of a larger effort by the university to transform its campuses into a real-world laboratory, tackling climate change head-on through innovative research and practical applications.
, director of Energy and Climate Action with University Services, Operations and Finance, highlights how these “living lab” projects align perfectly with the U of A’s strategic goals in renewables and sustainability.
Beyond funding Fleck’s work to realize wind energy potential on campus, Versteege’s team is also helping explore solar power potential. a professor in the Department of Civil and Environmental Engineering, has created a new AI and machine learning platform that uses Google Earth imagery to identify optimal locations for solar panels on campus buildings. This technology precisely maps out placement, considering shading, orientation and even existing rooftop infrastructure like ductwork to maximize energy generation while maintaining safety and aesthetics.
These seemingly separate projects are deeply intertwined. Both wind and solar modelling contribute to a comprehensive 3D model of the campus, allowing researchers to calibrate their computational models with real-world infrastructure.
Currently, the university boasts around two million watts of installed solar photovoltaic capacity, enough to power about 400 homes. The goal is to reach six to seven megawatts, optimizing energy savings during peak utility times and reducing greenhouse gas emissions.
“This holistic approach balances campus resiliency, the health and safety of students and staff, and significant energy savings,” says Versteege.
“These projects not only advance climate change solutions and energy reduction, but are a testament to how real-world research can translate into meaningful benefits for everyone.”