Leveraging Fine-Grained Occupancy Estimation Patterns for Effective HVAC Control

Background

One of the biggest culprits of energy consumption is HVAC. In 2017 alone, approximately 30% of energy consumption for commercial buildings in the U.S. was used for HVAC. Conventional uses for HVAC systems are not efficient at all and often lead to an excessive amount of unnecessary energy usage.

Usually, building operators use a static schedule for controlling HVAC systems without taking into account how many people use the building at different times of the day. Many HVAC systems operate at an assumed maximum occupancy in each room which isn't the case all the time. This can lead to significant energy waste, for example, an HVAC system providing ventilation for thirty people when there are only ten people in a room. Such widely used HVAC control designs miss opportunities to perform more accurate and efficient control.

Therefore there is a need for a more efficient process for controlling HVAC systems.

Technology

This technology revolves around using predicted occupant‑counts and accounting for misprediction costs to minimize energy consumption. This process will consist of first updating the thermal state of multiple zones in the building based on a building thermal model and information received from temperature sensors of the building. Then, the predicted occupant counts for an upcoming number of time slots for each of the zones in the building are updated by using the actual occupancy counts in those zones.

Finally, after these two variables are updated, there will be a misprediction type distribution that will be updated for the occupancy numbers in each of the zones for each of the time slots. This distribution will take into account true negatives, false positives, false negatives, and true positives which will lead to the total misprediction cost expectation according to the predicted occupant counts and the distribution.

This will result in determining HVAC power for each of the zones to optimize occupant thermal comfort (weighted according to the predicted occupant counts) while minimizing the total misprediction cost expectations.

Advantages

• More efficient than conventional HVAC systems.
• More cost‑effective as there is no unnecessary consumption of energy.
• Leads to a more comfortable environment for the occupants.

Application

This technology can be applied to a multitude of places due to the number of buildings that use HVAC systems. Almost every type of building can benefit from this technology.