Skip to main content
Skip to footer
Contact Us
We are taking you to another website now.
March 18, 2016

Office buildings around the globe consume large amounts of energy for air-conditioning to meet occupants’ comfort levels and ensure adequate indoor air quality. HVAC systems consume as much as 50% of the total energy in buildings. So it’s important to improve the operational efficiency of HVAC systems, to provide optimal comfort with minimal energy consumption. A couple of months ago, I spoke about centralized management of HVAC energy at BuildSys 2015. This post is based on that talk and the discussion at the event.

In order to cool or heat a space effectively, it is important to know if, and how many, people are occupying the building. According to the widely accepted standards set by the American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE), the exact occupant count is useful in catering to ventilation requirements. HVAC units typically use temporal variations in occupancy to adjust the temperature. During unoccupied durations, HVACs revert to setback temperature to reduce the cooling load on the Air Handling Unit (AHU), and thereby minimize energy consumption.

Often, a single large zone may be served by multiple AHUs, with multiple control knobs. In such a case, the return air temperature at any AHU depends on the temperature set-point on all the other AHUs. At any point in time, there can also be significant spatial variation in occupancy across the zone – for instance, in an airport, there may be more people congregated at the coffee counter and near the charging points.

Conventionally, each AHU is controlled independently using a Proportional-Integral-Derivative (PID) controller, which adjusts the flow rate to allow the air temperature to match the set point. This means that even if the occupants are only in one corner of a large room, the entire room will be cooled evenly, which makes for higher energy costs.

We believe it is possible to solve this problem with an intelligent control system. A coordinated control system can identify the flow rate of all AHUs together that can meet the desired comfort level with minimal use of energy, by solving the heat and humidity balance equations. The system should identify the optimum flow rates, accounting for the spatial occupancy variations in the room.

Such a solution can provide significant energy savings as compared to a traditional PID controller, especially when occupancy is not evenly distributed across the zone (such as in an office on a weekend). On the other hand, when the space is occupied more evenly (say during weekdays), the performance of the PID and the intelligent control system will be comparable.

By accurately sensing the temperature and occupancy of a given zone, it is possible for both the PID controllers and the intelligent control systems to save more energy when occupancy is skewed. However, this does not make much of a difference during weekdays, when occupancy is more even.

When it comes to real world occupancy patterns in office facilities, we find that reactive control is sufficient. A model predictive control (MPC) is required only for special occupancy patterns occurring in buildings such as conference halls, cinemas, and concert halls (where the occupancy is typically either high or zero).

What does this mean for you? Before deciding on an HVAC system for your building or space, make sure to properly identify the occupancy pattern. This will help you choose the right kind of AHU controls to ensure maximum energy savings.

Srinarayana Nagarathinam is a Senior Scientist and Lead Researcher for the biosafe and energy-smart building program at TCS Research. He holds a PhD from the Department of Mechanical Engineering, University of Sydney. Dr Nagarathinam is a recipient of the prestigious postdoctoral fellowship award by the Department of Science and Technology of the Australian government. He has extensive experience in thermal management of building systems, computational fluid dynamics, and heat transfer. His recent research interests include the application of physics-based and ML/AI-based techniques to optimal control of heating, ventilation, and air-conditioning (HVAC) systems.


Thank you for downloading

Your opinion counts! Let us know what you think by choosing one option below.