Identification of source location in a single-sided building with natural ventilation: Case of interunit pollutant dispersion

A sudden outbreak of COVID-19 occurred in December 2019 and its rapid spread over the last two years caused a global pandemic. A special airborne transmission via aerosols called interunit dispersion is risky in a high-density urban environment, which needs more attention. In order to identify the source location of pollutants or viruses under the interunit transmission condition with natural ventilation, this study adopted the inverse Computational Fluid Dynamics (CFD) simulation with the adjoint probability method. The detailed process of the inverse modeling was presented. Also, the possible interunit transmission routes of the pollutants or viruses were analyzed. A three-story building model with single-sided openings was built. Six different combinations of fixed sensor locations were tested, and it was determined that setting sensors in the four corner regions of the building was the optimist strategy. A total of 25 cases with five different wind directions (0 degrees, 45 degrees, 90 degrees, 135 degrees, and 180 degrees) were tested to verify the accuracy of the source location with inverse modeling. The results showed that 67%-78% of the rooms in the building can be identified with a limited number of pollutant sensors and all rooms can be identified with one additional sensor in the downstream room of the building under different wind direction. This research revealed that the inverse modeling method could be used to identify the pollutant source in the coupled indoor and outdoor environment. Further, this work can provide guidance for the pollutant monitor positions in the applications.

Automated summary

This study adopted inverse Computational Fluid Dynamics simulation and the adjoint probability method to identify the source location of pollutants or viruses under interunit transmission conditions. The results showed that a limited number of pollutant sensors could identify 67%-78% of the rooms in the building, and all rooms could be identified with one additional sensor in the downstream room under different wind directions.