MANDI: A team of researchers from Indian Institute of Technology Mandi in collaboration with a scientist from Université de Lorraine, France have developed a novel algorithm that automatically detects operational failures in Heating Ventilation and Air-Conditioning (HVAC) systems installed in buildings.
Detecting and identifying these faults manually in HVAC is a slow, costly, and error-prone process.
Delays in detecting and fixing the faults in HVAC systems can lead to indoor environmental issues, such as poor thermal comfort and indoor air quality, affecting occupant health and productivity.
This also results in the wastage of energy, thereby reducing building energy efficiency. Research suggests that faults in HVAC systems could increase the energy consumption of buildings by 4% to 18%.
But no more.
IIT Mandi researchers have developed an automated fault detection and diagnosis (FDD) algorithm that integrates with the building automation system (BAS) or building energy management system (BEMS).
This system helps maintenance crews identify and predict potential failures, analyze their impact, determine their importance to the BAS, and quickly direct repairs, ensuring that the system is always available.
Dr Tushar Jain, IIT Mandi said, “Our algorithm is robust against unmeasured disturbances and sensor noise, particularly the outdoor temperature, which affects the thermal dynamics of the building.”
The algorithm is strong enough to handle uncertainties such as the outside weather conditions and the number of occupants in the building.
This means that building engineers do not have to simplify the thermal dynamics model or use basic rule-based control or monitor algorithms that don't work well. The algorithm also estimates wall temperatures, which is important in predicting faults.
This algorithm is a testament to the innovative capabilities of IIT Mandi researchers and their commitment to advancing the building automation industry through cutting-edge technology.
The results of this research have been published in the Journal of Building Performance Simulation in a paper coauthored by Dr Tushar Jain, Associate Professor, School of Computing and Electrical Engineering and his research scholar Dr Mona Subramaniam from IIT Mandi and his collaborator Dr Joseph J. Yamé from Université de Lorraine, France
