Source characterization of airborne pollutant emissions by hybrid metaheuristic/gradient-based optimization...

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Computational Modeling

Background:

• Accurate identification of atmospheric contaminant sources is crucial for environmental protection and public health.
• Existing methods for source estimation often face challenges with complex atmospheric conditions and multiple emission points.

Purpose of the Study:

• To develop and validate a robust methodology for estimating single and multiple atmospheric emission sources.
• To enhance the accuracy and versatility of atmospheric dispersion modeling for source apportionment.

Main Methods:

• A hybrid approach combining metaheuristic and gradient-descent optimization techniques.
• Implementation of Tikhonov-type regularization for stable inverse problem solutions.
• Utilizing the Galerkin/Least-squares finite element formulation for realistic dispersion modeling.

Main Results:

• The proposed inversion model demonstrated high accuracy in identifying both single and multiple emission sources using experimental field data.
• Testing various configurations of regularization functionals and optimization techniques yielded consistent and reliable results.
• A discrepancy-based rule effectively determined the regularization parameter, ensuring optimal model performance.

Conclusions:

• The developed methodology provides a versatile and accurate tool for atmospheric emission source estimation.
• The approach offers competitive computational efficiency, making it suitable for practical applications.
• The study validates the effectiveness of hybrid optimization and regularization in complex environmental modeling scenarios.