Simulação de uma Pluma de Contaminantes no Rio Macaé Utilizando Redes Neurais Artificiais

Content Provider	Semantic Scholar
Author	Telles, Wagner Rambaldi Rodrigues, Pedro Paulo Gomes Watts Neto, Antônio José Da Silva Santiago, Orestes Llanes Mesa, Mirtha Irizar Junior, Jader Lugon
Copyright Year	2013
Abstract	The risk of a lack of drinking water is a subject of worldwide debate. Some countries already suffer from this problem due to uncontrolled water pollution by industrial and domestic waste, among others. Based on this, mathematical and computational models have been constantly used in the study of contaminant transport in river systems. They are important in the prevention and estimation of the behavior of contaminant plumes in these water bodies. This paper is a first approach to modeling the behavior of contaminants in Macae River, a major river located in the northeast part of the Brazilian state of Rio de Janeiro. It represents a starting point for studies based on Artificial Neural Networks, in lieu of a mathematical model of 2D transport of pollutants. As a first approach, the behavior of a plume of contaminants is simulated for a stretch of Macae River. Experimental data are compared to the results obtained by neural networks using LevenbergMarquardt as the training method. The results presented in this work will serve as a basis for subsequent training of the neural network in order to estimate conservative substance Simulação de uma Pluma de Contaminantes no Rio Macaé Utilizando Redes Neurais Artificiais 174 concentrations at any point in a given stretch of Macae River. Key-words: Macae River, pollutant transport, artificial neural networks.
Starting Page	165
Ending Page	174
Page Count	10
File Format	PDF HTM / HTML
DOI	10.21168/rbrh.v18n2.p165-174
Volume Number	18
Alternate Webpage(s)	https://abrh.s3.sa-east-1.amazonaws.com/Sumarios/98/67ba98f2f7dff990d90418270ad7ee68_349db00e06eb1c1879ec2cacf345eba5.pdf
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article