Promoting Energy Efficiency of HVAC Operation in Large Office Spaces with a Wi-Fi Probe enabled Markov Time Window Occupancy Detection Approach

Abstract

In recent years, demand-based control in HVAC systems have achieved a great amount energy saving by providing precise services in according to actual demand. As the premise of determining actual demand, occupancy information has a significant impact on building heating, cooling and ventilation in building control and energy auditing. In this paper, a Wi-Fi probe based occupancy detection method is applied to detect the occupancy in typical office buildings. Given the Wi-Fi coverage and smart devices are widely used in modern buildings, Wi-Fi probe requires no initial investment and could scan the Internet connection request and response of occupants. Previous studies suggest time-series and stochastic characteristics of occupancy information, this research proposed a Time-Window based Markov Chain (TWMC) model to detect occupancy. An on-site experiment was conducted in this study to validate the proposed method. The results report an accuracy of over 80% (x-accuracy when x equals 4). Compared to actual occupancy profile, the proposed model shows over 88% of supply air amount reduction in energy simulation with an absolute deviation less than 20%. By integrating the proposed TWMC model based on Wi-Fi probe and demand-based control system, the energy consumption of HVAC system could be significantly reduced.