BIM-supported sensor placement optimization based on genetic algorithm for multi-zone thermal comfort and IAQ monitoring

Abstract

Thermal comfort and indoor air quality (IAQ) are important to occupants’ wellbeing. Sensors are deployed for pollutant exposure assessments and long-term IAQ monitoring. How to optimize sensor placements in single rooms has been studied. However, limited studies on sensor placement in a multi-zone indoor environment have been conducted. On a typical office floor, room partitions can be changed by tenants, and conducting field measurements in the occupied tenant areas is not allowed. It is difficult to determine where to install a sensor. The related studies rarely consider different weather conditions and ventilation system settings when performing optimization. This study aims to develop a methodology to optimize temperature and CO2 sensor placement for thermal comfort and IAQ monitoring in a multi-zone environment under limited field measurement. Supported by building information modeling (BIM) technology, different seasons, MVAC system settings, and occupancy scenarios were taken into account when optimizing sensor placement. How to overcome limited information of the inaccessible regions by applying the conservation equations as well as considering the MVAC layout of the floor and sources of heat and CO2 emission were also illustrated. The methodology was illustrated using a validated Computational fluid dynamics (CFD) model of a typical office floor. With the use of genetic algorithm (GA) and machine learning, the generated sensor placements fulfilled coverage requirements from LEED and over 70% of the regions with significant fluctuation in temperature and CO2 are inside the sensing range of sensors. The developed approach can be applied in both new and existing buildings.