Real-Time Occupancy Detection and Integration with HVAC Control Systems in Buildings Considering Energy Saving and Thermal Comfort

With advancements in Internet of Things (IoT) technologies, integrating occupancy information into building energy management systems and services including heating, ventilation, and air conditioning (HVAC) systems and lighting has evolved beyond traditional sensor-based methods. The integration of IoT in HVAC systems offers a promising approach to analyze and enhance energy efficiency. Because building services are primarily designed to ensure satisfactory indoor conditions for occupants, accurately determining real-time occupancy within building spaces is crucial for optimal control. This paper evaluates the performance of two occupancy detection methods: computer vision (CV) techniques and wireless sensors. The better the real-time occupancy detection, the more efficient the HVAC control will be. We apply a CV technique to an occupancy-based control (OBC) algorithm tailored for a single-zone office building and evaluate the performance by analyzing zone thermal comfort using metrics such as Predicted Mean Vote (PMV), Percentage of People Dissatisfied (PPD), and direct occupant feedback. Experimental results for occupancy detection indicate that the CV method achieves an accuracy of 98%, compared to 85.1% using wireless sensors. In addition, the OBC system using the CV method reduces energy consumption by 54.68% while maintaining an overall thermal comfort satisfaction of 82.5%.