Express Computer
Home  »  Artificial Intelligence AI  »  IIT Mandi designs algorithms for HVAC systems in buildings

IIT Mandi designs algorithms for HVAC systems in buildings

0 233

The Indian Institute of Technology Mandi (IIT Mandi) on has said its researchers have developed new algorithms for component failure detection and diagnosis that can enhance the energy efficient operation of Heating, Ventilation and Air Conditioning (HVAC) systems used in buildings.

In centralised HVAC systems in buildings, climate control and ventilation are performed at a centralised location outside the building by an Air Handling Unit (AHU) which results in better maintenance and no indoor noise.

The processed air is distributed to every room with the help of controlled ducts and excess air in the room is recirculated through the unit.

The effective operation of centralised HVAC systems requires the careful orchestration of the various components. A faulty component may hamper the efficient operation of HVAC and increase the costs of operation, the institute said.

Furthermore, a faulty component increases the load on the other healthy components, thereby increasing wear and tear and reducing the life of the entire system.

“Variable-air-volume (VAV) terminal boxes are an important component of centralized HVAC systems. Any faults or failures in these VAV boxes can drastically affect the control performance,” said Tushar Jain, Assistant Professor, School of Computing and Electrical Engineering, IIT Mandi.

“The VAV dampers play a significant role in the supervisory health-aware control strategy of the system, and timely and automatic detection of faults in these components can be very useful in the management of the health of the HVAC”, Jain explained.

According to the researchers, the algorithm developed for detection and estimation of the magnitude of the failure of VAV dampers uses analytical models that are applicable over a wide range of unpredictable operating conditions, such as weather dynamics, outside air temperature, zone occupancy profile, and so on.

“The wall temperature, which is usually ignored in climate control, is an important parameter for efficient function of the HVAC, and our algorithm takes this into consideration”, he added.

The researchers have demonstrated the effectiveness through exhaustive simulation studies and have shown that the developed algorithm can successfully detect and estimate the magnitude of VAV multiple damper faults.

The research team is working towards developing decentralised and distributed fault diagnosis algorithms and fault-tolerant control strategies for large scale buildings in order to ensure more energy-efficient operation of HVAC systems and hopes to extend this work to real time testing and validation on a real building monitoring platform.

The results of the team’s recent work have been published in the Journal of Building Engineering, Elsevier.

Get real time updates directly on you device, subscribe now.

Leave A Reply

Your email address will not be published.

LIVE Webinar

Digitize your HR practice with extensions to success factors

Join us for a virtual meeting on how organizations can use these extensions to not just provide a better experience to its’ employees, but also to significantly improve the efficiency of the HR processes
REGISTER NOW 
India's Leading e-Governance Summit is here!!! Attend and Know more.
Register Now!
close-image
Attend Webinar & Enhance Your Organisation's Digital Experience.
Register Now
close-image
Enable A Truly Seamless & Secure Workplace.
Register Now
close-image
Attend Inida's Largest BFSI Technology Conclave!
Register Now
close-image
Know how to protect your company in digital era.
Register Now
close-image
Protect Your Critical Assets From Well-Organized Hackers
Register Now
close-image
Find Solutions to Maintain Productivity
Register Now
close-image
Live Webinar : Improve customer experience with Voice Bots
Register Now
close-image
Live Event: Technology Day- Kerala, E- Governance Champions Awards
Register Now
close-image
Virtual Conference : Learn to Automate complex Business Processes
Register Now
close-image