Eventos Anais de eventos
ENCIT 2020
18th Brazilian Congress of Thermal Sciences and Engineering
GENETIC ALGORITHM OPTIMIZATION APPLIED TO SOURCE ESTIMATION IN THE ATMOSPHERE
Submission Author:
Roseane Albani , RJ
Co-Authors:
Roseane Albani, Vinicius Albani, Antônio Silva Neto
Presenter: Roseane Albani
doi://10.26678/ABCM.ENCIT2020.CIT20-0160
Abstract
This work proposes a new methodology to estimate atmospheric emissions using a combination of composite misfit functions in the Tikhonov-type regularization and the Genetic Algorithm, to minimize this functional. A Morozovlike discrepancy principle is used to choose the regularization parameter according to the available data. The joint use of composite misfit functions, Tikhonov-type regularization, and the Genetic Algorithm defines the inverse modeling for source identification. In addition, the dispersion model (or forward problem) is solved using the Galerkin/Least-Squares formulation combined with parametric profiles to account for the relevant process in the atmospheric dispersion considered in this work. The proposed methodology is evaluated with a single source dispersion field experiment, presenting accurate values for the source position and strength.
Keywords
genetic algorithm, atmospheric dispersion, Source Estimation, Tikhonov regularization, Composite Misfit
DOWNLOAD PDF VIEW PRESENTATION
