Leak detection for UBC buildings using autonomous robots

Project Team

Zhengbo Zou, Assistant Professor, Civil Engineering/Applied Science

Bryan Archibald, Manager of Maintenance Planning, UBC Building Operations

Buildings account for 25% of the total energy consumed in Canada and 12% of the greenhouse gas (GHG) emissions¹. This project aims to address the issue of high energy consumption and greenhouse gas emissions in buildings by improving the process for detecting defects in heating, ventilation, and air conditioning (HVAC) systems and enabling faster and easier repairs

Researchers in the Leak Detection project will improve and test a developed robotic solution that can automatically navigate tight spaces, generate a precise map of the environment, and detect leaks in HVAC systems using onboard sensors and machine learning algorithms. Currently, HVAC system inspections are manual and reactive, which are time consuming and often only occur after buildings have been damaged from system leaks.

Large building portfolio managers like UBC have a significant opportunity to become proactive in addressing HVAC system leaks, blockages and other damages, in existing buildings on campus. These defects can increase the energy consumption of buildings by 20% for cooling climates and 30% for heating climates^₂. The project will work with UBC building operations staff to identify test buildings for robot deployment and aims to test the robot in 5 buildings with diverse designs, ages, and functions.

Piloting the deployment of the developed robots³ in UBC buildings will serve as a proof-of-concept for automated HVAC inspection and leak detection. The Leak Detection project will contribute to UBC’s commitment to reduce building energy consumption and GHG emissions as set out in the Climate Action Plan 2030. The diversity of the test buildings will provide valuable data for the robotic solution's scalability to various building types and increase potential application beyond campus⁴. This project provides a creative solution to HVAC leak detection for the vast majority of existing buildings that do not have smart HVAC systems. 

_{¹Canada Energy Regulator. (2022). “Provincial and Territorial Energy Profiles”}

_{²Fisk, W. J., Delp, W., Diamond, R., Dickerhoff, D., Levinson, R., Modera, M., ... & Wang, D. (2000). Duct systems in large commercial buildings: physical characterization, air leakage, and heat conduction gains. Energy and Buildings, 32(1), 109-119.}

_{³Cai, W., Huang, L., & Zou, Z. (2023, November). RoboAuditor: Goal-Oriented Robotic System for Assessing Energy-intensive Indoor Appliance via Visual Language Models. In Proceedings of the 10th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (pp. 130-139).}

_{⁴Duan, K., Suen, C. W. K., & Zou, Z. (2023). Robot morphology evolution for automated HVAC system inspections using graph heuristic search and reinforcement learning. Automation in Construction, 153, 104956.}