Fast Identification of Multiple Indoor Constant Contaminant Sources by Ideal Sensors: A Theoretical Model and Numerical Validation

Abstract

If hazardous contaminants are suddenly released indoors, quickly identifying the pollutants characteristics, such as the locations and emission rates, of the contaminant sources is critical for developing fast and effective response measures. This study presents a theoretical model for quickly identifying the locations and emission rates of multiple constant sources indoors using a single or limited number of ideal sensors. The model combines a linear programming model with an analytical expression of indoor contaminant dispersion that was presented in our previous study. Before the release of contaminants, only a limited number of time-consuming computational fluid dynamics (CFD) simulations need to be conducted to cover a large number of possible scenarios because of the analytical expression integrated into the model. After the release of contaminants, the model can be solved in real-time. Through case studies of sixteen contaminant release scenarios in a three-dimensional office, the effectiveness of the model was numerically demonstrated and validated. The results revealed that the identification accuracy of the presented model was closely related to the sensor layouts and the total sampling time, rather than the sampling interval.