Identifying spatiotemporal information of the point pollutant source indoors based on the adjoint-regularization...

Area of Science:

• Environmental Science
• Public Health
• Computational Fluid Dynamics (CFD)

Background:

• Accurate identification of indoor pollutant sources is crucial for public health and safety.
• Existing methods struggle with temporal source identification and spatial analysis of multiple sources.

Purpose of the Study:

• To develop a novel algorithm for spatiotemporal identification of point pollutant sources in indoor environments.
• To improve upon limitations of current adjoint probability and optimization methods.

Main Methods:

• A hybrid approach combining adjoint-pulse and regularization methods was developed.
• Source-receptor response matrices were generated using a validated CFD model.
• Composite Bayesian inference was employed to determine release rate and location.

Main Results:

• The algorithm achieved mean absolute percentage errors (MAPEs) below 15% for source intensity estimation.
• Source localization success rates exceeded 25 out of 30 trials.
• The method demonstrated effectiveness in identifying dynamic pollutant sources.

Conclusions:

• The proposed algorithm offers a robust solution for identifying indoor pollutant source spatiotemporal information.
• This technique holds significant potential for detecting airborne pathogen sources in public buildings.
• Integration with disease-specific biomarker sensors can further enhance public health surveillance.