Improving the tools for assessment and analysis of energy losses through building envelopes using Infrared Thermography (IRT), Unmanned Aerial Systems (UAV), and Artificial Intelligence (AI)

IR imaging presents the temperature distribution of the exterior surface of a wall and is typically assessed through visual inspection by building science experts. A specialist must review numerous images one by one which is inefficient and inaccurate. The use of Unmanned Aerial Vehicles (UAVs) with IR camera attachments has become a method of faster and more accessible on-site building envelope evaluation. The proposed research aims to develop a machine learning framework to improve the evaluation of energy loss through the envelope, using UAV and IR thermography. An artificial intelligence algorithm will help significantly cut down assessment times and make the process more advanced. Understanding the heat transfer process within the wall system without a need for intrusive openings will allow for representative and more applicable retrofit intervention strategies. This research and development will drive the building rehabilitation industry to be more practical and effective.

Faculty Supervisor:

Miljana Horvat;Umberto Beradi

Student:

David Gertsvolf

Partner:

QEA Tech Inc

Discipline:

Architecture and design

Sector:

Professional, scientific and technical services

University:

Ryerson University