CO2 based occupancy detection algorithm: Experimental analysis and validation for office and residential buildings

Abstract The detection of occupants in indoors can be fundamental for a correct operation of the installed engineering systems (e.g. lighting, ventilation, heating and cooling). Further, real occupancy profiles can be used as input of stochastic models for dynamic simulation of buildings and their engineering systems. In this work an algorithm for the detection of occupants in the indoor environment is presented, validated and evaluated among different scenarios. The algorithm is based on the concentration of carbon dioxide in the indoor air. The testing and validation has been done both for residential and non-residential buildings: two offices with mechanical ventilation system, one office without mechanical ventilation, a kitchen and a big sleeping/living room of a residential building without mechanical ventilation have been evaluated. Volunteers recorded their presence profiles in the monitored rooms to permit the validation of the algorithms. The results of the algorithms for the detection of occupants (whether occupants are present or not) provides correct presence profile up to 95.8% of the time while the exact number of occupants in the rooms is correctly identified up to 80.6% of the time.