Journal of Korean Institute of Architectural Sustainable Environment and Building Systems ISSN:1976-6483(Print) 2586-0666(Online) 한국건축친환경설비학회논문집

IoT sensor data that impose occupancy information could be very useful in building energy simulation tools for improving accuracy of the simulation result. By doing so, those tools can be embedded in BEMS (building energy management system) for more strategic and efficient building energy management. This study focuses on the development of a model predictive control software environment with the IoT sensor data. As a demonstration, IoT sensor data is gathered from private office space. Occupancy estimation is implemented through classification algorithm (decision tree) with the data gathered. The actual but estimated occupancy information is transferred to EnergyPlus via BCVTB. Lighting energy consumption is predicted from the MPC simulation and compared with measured lighting energy data. It was found that this MPC environment could save lighting energy when the occupant leave the space without turn off the light. Around 25% of wasted lighting energy was saved by turning off the light when the space is vacancy. This demonstration shows the promising aspect of estimation of occupancy and MPC technology.