Launch of the UTMOST FDD Project
Heating, ventilation and air conditioning systems operate silently in the background of modern buildings, ensuring thermal comfort and healthy indoor air. Yet when faults occur in these systems, they often remain unnoticed for long periods, leading to increased energy consumption, higher operating costs and degraded indoor air quality.
A new research initiative called PRIN 2022 MUR - UTMOST FDD has been launched to address this challenge. The project brings together researchers from Politecnico di Torino and the Università della Campania with the goal of developing innovative tools capable of automatically detecting operational faults in air handling units (AHUs), one of the key components of building ventilation systems.
Fault Detection and Diagnosis (FDD) technologies have been widely investigated in recent years, but their practical application still faces several limitations. Many existing solutions rely exclusively on either data-driven algorithms or expert knowledge. These approaches often struggle to generalize across different HVAC systems and typically require large amounts of labelled data.
The PRIN 2022 MUR - UTMOST FDD project aims to overcome these limitations by developing hybrid diagnostic strategies that combine machine learning methods with knowledge-based rules derived from HVAC expertise. By merging these two perspectives, the project seeks to create more robust and transferable diagnostic tools.
The research activities will focus on several key areas. First, an extensive experimental dataset will be generated using a fully instrumented air handling unit operating under both normal and faulty conditions. Researchers will then develop a digital twin of the system capable of reproducing its dynamic behaviour and simulating a wide range of operational scenarios. Finally, advanced diagnostic algorithms will be designed and tested using both experimental and simulated data.
One of the distinctive aspects of the project is its commitment to open science. The experimental dataset, the digital twin simulation results will be made publicly available, allowing researchers and industry professionals to benchmark and improve their own diagnostic solutions.
By improving the reliability and transferability of FDD technologies, the project aims to support the development of smarter building energy management systems and contribute to more efficient and sustainable building operation.
Stay tuned for the next updates, where we will share the first results of the experimental activities and provide insights into how digital twin models and artificial intelligence can help uncover hidden faults in HVAC systems.
Acknowledgements
The project is funded by the European Union in the framework of the initiatives "Next Generation EU"
