ΑΙhub.org
 

Machine learning-aided thermography for building heat loss detection


by
21 June 2024



share this:

Image credit: Passivhaus Institut, CC BY-SA 3.0 , via Wikimedia Commons.

University of Waterloo researchers have developed a new method that could lead to significant energy savings in buildings. The team identified 28 major heat loss regions in a multi-unit residential building with the most severe ones being at wall intersections and around windows.

Building enclosures rely on heat and moisture control to avoid significant energy loss due to airflow leakage, which makes buildings less comfortable and more costly to maintain. This problem will likely be compounded by climate change due to volatile temperature fluctuations. Since manual inspection is time-consuming and infrequently done due to a lack of trained personnel, energy inefficiency becomes a widespread problem for buildings.  

Researchers at Waterloo created an autonomous, real-time platform to make buildings more energy efficient. The platform combines a deep learning method, infrared technology, and a mathematical model that quantifies heat flow to better identify areas of heat loss in buildings.

Using the new method, the researchers conducted an advanced study on a multi-unit residential building in the extreme climate of Canadian prairies, where elderly residents reported discomfort and higher electricity bills due to increased demand for heating in their units. Using machine learning tools, the team trained the program to examine thermal images in real time, achieving 81 percent accuracy in detecting regions of heat loss in the building envelope.  

“The almost 10 per cent increase in accuracy with this AI-based model is impactful, as it enhances occupants’ comfort as well as reduces energy bills,” said Dr. Mohamad Araji, director of Waterloo’s Architectural Engineering Program and head of the Symbiosis Lab, an interdisciplinary group at the university that specializes in developing innovative building systems and building more environmentally friendly buildings.  

The new AI tools helped to remove the element of human error in examining the results and increased the speed of getting the data analyzed by a factor of 12 compared to traditional building inspection methods. 

Future expansions to this work will include utilizing drones equipped with cameras to inspect high-rise buildings. 

“The hope is that our methodology can be used to analyze buildings and lead to millions in energy savings in a much faster way than previously possible,” Araji said.  

Reference

Machine learning-aided thermography for autonomous heat loss detection in buildings, Ali Waqas and Mohamad T. Araji, Energy Conversion and Management, 2024.



tags: ,


University of Waterloo




            AIhub is supported by:



Related posts :



Congratulations to the #AIES2025 best paper award winners!

  21 Oct 2025
The four winners of best paper prizes were announced during the opening ceremony at AIES.

From the telegraph to AI, our communications systems have always had hidden environmental costs

  20 Oct 2025
Drawing parallels between new technologies of the past and today.

What’s on the programme at #AIES2025?

  17 Oct 2025
The conference on AI, ethics, and society will take place in Madrid from 20-22 October.

Generative AI model maps how a new antibiotic targets gut bacteria

  16 Oct 2025
Researchers used a GenAI model to reveal how a narrow-spectrum antibiotic attacks disease-causing bacteria.

What’s coming up at #IROS2025?

  15 Oct 2025
Find out what the International Conference on Intelligent Robots and Systems has in store.

Applying machine learning to chip design and manufacturing: interview with Lorenzo Servadei

  14 Oct 2025
Find out how Lorenzo and his team are using ML and Electronic Design Automation.

Why we should be skeptical of the hasty global push to test 15-year-olds’ AI literacy in 2029

  13 Oct 2025
Are schools set to become testing grounds for AI developments?

Machine learning for atomic-scale simulations: balancing speed and physical laws

How much underlying physics can we safely “shortcut” without breaking a simulation?



 

AIhub is supported by:






 












©2025.05 - Association for the Understanding of Artificial Intelligence