ABSTRACT
The changing earth’s climate is a global concern, and nations around the world have started several initiatives to mitigate its impacts. In addition to reducing greenhouse gas emissions, it is necessary to improve energy efficiency by reducing the energy consumed in households, industries, and other commercial buildings. Buildings equipped with Heating, Ventilation, and Air Conditioning (HVAC) system need to control unnecessary energy consumptions that might occur due to forgetting devices turned on, which necessitates the need to tune the devices automatically. Smart sensing together with machine learning can help this cause by shutting off HVAC to unoccupied rooms and by automatic tuning according to user preferences to improve user satisfaction. Such a smart building usually implements a nonintrusive, low-cost multimodal sensor agent for motion detection, sound level, lighting conditions, CO2 level, and door state sensing. The Internet of Things (IoT), which can collect and monitor a large amount of data on different aspects of a building, feeds the data to the processor of Smart Building Energy Management system (SBEMS). The massive volume of sensory data collected from sensors needs to be analyzed by machine learning algorithms and mined to accurately estimate the number of occupants in each room and to predict future occupancy to facilitate timely actions and better decision-making. By using these intelligent machine models to predict room occupancy in advance, the smart buildings can adjust set points and schedules of HVAC system accordingly to minimize energy consumption while maintaining occupant comfort. This chapter aims to describe the data captured by IoT sensors, data preprocessing, feature set, the role of machine learning and Big Data analytics for such smart service in SBEMS.
