Determinação de uma fonte de contaminantes no estuário do Rio Macaé através da técnica de problemas inversos

• Autor(a) principal: Parolin, Radael de Souza
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/13664
• topic: Pollutant transport; Determination of sources; Inverse problem; Finite Element Method; Finite Difference Method; Transporte de poluentes; Determinação de fontes; Método dos Elementos Finitos; Método de Diferenças Finitas; Hidrodinamica Modelos matemáticos; Problemas inversos (Equações diferenciais); Rio Macaé (RJ) Estuários; CNPQ::CIENCIAS EXATAS E DA TERRA::MATEMATICA::MATEMATICA APLICADA

A great challenge of our time is the preservation of water resources, as well as the proper management of the same, given the expansion of cities and human activities. The quality of a water body is usually evaluated through the analysis of biological, physical and chemical parameters. The behavior of these parameters can conveniently be simulated by mathematical and computational models, which arise as a useful tool by its ability to generate scenarios that may support, for example, decision making. In this thesis we discuss techniques for estimating the location and intensity of a source of conservative contaminant hypothetically released in the predominantly fluvial area of an estuary. The release here considered happens in a punctual and continuous form and the focused area includes the River Macaé estuary, located in the northern coast of Rio de Janeiro. The work includes the solution of a direct problem, which consists of two-dimensional transport (vertically integrated) of that hypothetical contaminant, as well as the application of inverse problems techniques. For the solution of the contaminant transport, modeled here by the version 2D horizontal advectiondiffusion equation, were used as discretization methods the Finite Element Method and the Finite Difference Method. For the hydrodynamic problem were used data from a solution already developed to the River Macaé estuary. Analyzed the mesh according to the discretization method, were defined the geometry of the estuary and the hydrodynamic and transport parameters. For estimation of the proposed parameters the technique of inverse problems was used, with the use of the Luus-Jaakola method, the Particle Collision Algorithm and the Ant Colony Optimization for estimating the location and of the Golden Section method for estimating the parameter of intensity of the source. For the definition of a source with the objective of proposing an idealized experimental scenery and of the collection of sampling data the sensibility analysis was accomplished with relationship to the parameters to be estimated. As sampling data of concentration were synthetic data, the inverse problem was solved using them with and without noise, introduced artificially and randomly. Without the noise use, the three methods were shown equally efficient, with accurate estimation in 95% of the executions. Already using sampling data with noise of 5%, the Luus-Jaakola method was shown more efficient in effort and computational cost, although all have estimated the source precisely in 80% of the executions. Considering the reached results in this work it is possible to estimate a source of constituents, as to its location and intensity, through the technique of inverse problems. Moreover, the applied methods were shown efficient in the estimation of such parameters with accurate estimates for most executions. Thus, the study of the behavior of contaminants, and especially the identification of external sources, it becomes an important tool for the management of the water resources, making possible, besides, the identification of possible responsible for environmental liabilities.