Occupancy estimation using environmental parameters

Abstract

Energy consumption in Singapore has been rising in recent years. A huge contributor to this trend comes from heating, ventilation, and air conditioning (HVAC) systems in modern buildings, where energy may be wasted to provide cooling unnecessarily. As a result, energy-saving technologies are being studied and introduced in Singapore, to slow down the growth of electricity consumption and reduce electricity wastage. One such study field involves the prediction of occupancy levels, by incorporating data retrieved from environment sensors, with machine learning techniques. This paper thus covers the analysis of several measured environmental parameters, combined with some machine learning models, to effectively produce occupancy statuses of an indoor environment. Moreover, the machine learning models utilised will be evaluated and discussed, to identify the suitable models to apply for the conservation of energy consumption, for relevant electrical systems and appliances in buildings.