Team Projects

Team:

Lawrence Technological University

Name of the energy conservation measure (ECM):

Universal Artificial Intelligence (AI)-based occupant comfort control framework

Purpose of ECM:

Today, we have an unprecedented opportunity to restore what we have built. The total world building stock is 150 billion m2 and it is estimated that over 80 billion m2 of buildings will be newly built or retrofitted in urban areas worldwide. That equals roughly 3.5 times the total U.S. building stock that can be transformed into energy-efficient or carbon-neutral buildings. If these changes occur, it will tremendously reduce overall Greenhouse Gas (GHG) emissions from the building industry.

Retrofitting existing buildings has been a critical pathway to pursue energy-efficient or carbon-neutral buildings. Several retrofit strategies have been explored and a building automation system (BAS) -- centralized automatic control of a building’s indoor conditions to maintain occupant comfort -- is one of the most promising strategies.

Description of ECM:

The research team proposed a novel Universal Artificial Intelligence (AI)-based occupant comfort control framework -- a “Plug and Play” technology that does not require in-depth knowledge because the framework will learn and calibrate the user’s comfort preference automatically with machine learning algorithms. The framework offers an “All-in-One” data acquisition system by replacing all other sensors with universal infrared (IR) array sensors.

The IR array sensor reads a grid of surface temperatures, and these data are used to infer information such as the number of occupants in a room, their activity level, indoor surface temperatures, irradiation, and illuminance levels.

Additionally, the team’s control framework can adjust to changing conditions in the room. In this adaptive control, the system algorithms allow controllers to track changes that are normally described as the ‘changing parameters’ within the control system. The proposed framework will be able to control all devices that are influencing the thermal comfort level of the space with universal occupancy sensing and building control utilizing IR array sensors and machine learning algorithms.

The goal is to accurately predict occupancy and fine-tuned control of BAS.