Computer-vision-assisted subzone-level demand-controlled ventilation with fast occupancy adaptation for large open spaces towards balanced IAQ and energy performance

It is essential to retrofit the conventional constant air volume air-conditioning of large open spaces with improved demand-controlled ventilation (DCV) for balanced indoor air quality and energy consumption in the building sector. However, existing improved DCV strategies do not effectively distinguish intensively occupied and less occupied zones and fail to achieve on-demand ventilation of individual subzones. This study thus employes the computer-vision-assisted occupancy detection method and proposes a systemic solution for large open spaces, which enable subzone-level demand-controlled ventilation that can perform fast response to the dynamic occupancy profile of each subzone. The objective is to enhance the subzone-level environment quality with balanced performance of IAQ, thermal comfort and energy consumption. The proposed control strategy has been evaluated in a simulated large open classroom. The results show that, compared to existing direct–CO₂–based DCV control strategy, up to 12.22% lower CO₂ concentration (1154 ppm by direct–CO₂–based DCV, while 1013 ppm by proposed strategy) of occupied subzones can be realized while avoiding unnecessary ventilation for less occupied or unoccupied subzones. When combined with the PMV-based indoor thermal comfort control, the proposed strategy shows cooling load reduction potentials by maximal 49.4% and electricity energy saving by maximal 33.52% when compared to fixed fresh air strategy, and by maximal 15.57% of cooling load reduction as well as by maximal 8.75% of electricity energy saving when compared to direct–CO₂–based DCV control strategy. © 2023 Elsevier Ltd