Agent based HVAC optimization model for building energy efficiency

The Heating, Ventilation and Air Conditioning (HVAC) systems contribute to a considerable percentage (about 50%) of a building´s energy consumption. So controlling them will have a significant impact on the energy efficiency of the building, thereby reducing the carbon footprint. The factors that affect the indoor conditions include static factors like the architecture of the building, the number, size and orientation of the rooms and the windows and doors in them, and dynamic factors like the outside weather and the number of occupants in the room. In most of the implementations and proposals for HVAC control automation, the occupancy is measured with the help of one or more sensors like CO₂ sensors, Passive Infrared Sensors, Acoustic Sensors, Magnetic reed Sensors, video based inputs, etc. These systems require the sensors to be placed and monitored at frequent intervals to get real time occupancy estimation, which involves significant performance overhead. For buildings like educational institutions or offices with fixed time slots, where the seating arrangements of occupants are predictable based on available information, the punching data can be used instead for estimating the occupancy of rooms. Since the punching data is already available as most of these offices and institutions use them to keep record of attendance, and is easily accessible, this system can be easily implemented in the buildings which does not have built-in occupancy measuring devices. The HVAC control based on this system designed using an agent based model is proposed. Probability assessment is also done by the agents to predict the exact occupancy based on previous data, to make the estimation more realistic.