Aplicação de um Comitê de Redes Neurais Artificiais para a Solução de Problemas Inversos em Transferência Radiativa
DOI:
https://doi.org/10.5540/tema.2010.011.02.0171Abstract
Este trabalho fundamenta-se no conceito de Máquina de Comitê de Redes Neurais Artificiais e tem por objetivo resolver o problema inverso de transferência radiativa em um meio unidimensional, homogêneo, absorvedor e espalhador isotrópico. A Máquina de Comitê de Redes Neurais Artificiais agrega e combina o conhecimento adquirido de um certo número de especialistas aqui representados, individualmente, por cada uma das Redes Neurais Artificiais (RNAs) que compõem o Comitê. O objetivo é atingir um resultado final hipoteticamente melhor que o obtido por qualquer rede neural especialista separadamente. O uso desta técnica pode reduzir o desperdício computacional que ocorre ao se treinar exaustivamente várias RNAs, separadamente, selecionando-se apenas a rede que apresente a melhor generalização e descartando-se as demais. Neste trabalho são obtidas, usando-se a técnica de Máquina de Comitê de Redes Neurais Artificiais, estimativas de parâmetros de transferência radiativa: espessura óptica, albedo de espalhamento simples e reflectividades difusas do meio participante sob análise. Finalmente, os resultados obtidos são comparados com os encontrados usando-se redes Perceptrons de Múltiplas Camadas (PMCs) individualmente, denominadas neste trabalho como redes especialistas e mostrando que a técnica empregada traz significativasReferences
[1] W.J. Blackwell, A neural-network technique for retrieval of atmospheric temperature
and moisture profiles from high spectral resolution sounding data, IEEE Trans Geosci Remote Sens, 43 (2005), 2535–2546.
[2] H.F. de Campos Velho, M.R. Rematoso, M.T. Vilhena, Inverse problems forestimating bottom boundary conditions of natural waters, Int J Numer Meth Engng, 54(9)(2002), 1357–1368.
[3] H.F. de Campos Velho, J.D.S. Silva, E.H. Shiguemori, Hardware implementation for the atmospheric temperature retrieval from satellite data, Inverse Problems, Design and Optimization Symposium (IPDO) 1,(2007), 349–353.
[4] E.S. Chalhoub, H.F. de Campos Velho, Multispectral reconstruction of bioluminescence term in natural waters, Appl Numer Math, 47 (2003), 365–376.
[5] S. Chandrasekhar, “Radiative Transfer”, Dover Publications Inc., New York, 1960.
[6] S. Haykin , “Neural Networks: A Comprehensive Foundation”, Prentice Hall, 1994.
[7] R.C. Oliveira, N.I. Alvarez Acevedo, A.J. Silva Neto, L. Biondi Neto, Aplicação da técnica de máquina de comitê de redes neurais artificiais para a solução de problemas inversos em transferência radiativa, XI Encontro de Modelagem Computacional, Volta Redonda, R.J., (2008).
[8] M.N. Özisik, “Radiative Transfer and Interactions with Conduction and Convection”, John Wiley & Sons, USA, (1973).
[9] M. Riedmiller, H. Braun, A direct adaptative method for faster backpropagation learning: the RPROP algorithm Proc. IEEE International Conference on Neural Networks, (1993), 234–241.
[10] A.J. Silva Neto, F.J.C.P. Soeiro, Inverse Problem of Space Dependent Albedo Estimation with Artificial Neural Networks and Hybrid Methods, Proc. 18th International Congress of Mechanical Engineering, COBEM, ABCM, Ouro Preto, Brazil, (2005).
[11] A.J. Silva Neto, F.J.C.P. Soeiro, H.F. Campos Velho, P. Oliva Soares, Using Neural Networks to Obtain Initial Estimates for the Solution of Inverse Heat Transfer Problems, Inverse Problems, Design and Optimization Symposium, Rio de Janeiro, Brazil,(2004).
[12] F.J.C.P. Soeiro, A.J. Silva Neto, Solution of Inverse Radiative Transfer Problems in Two-Layer Media with Artificial Neural Networks, 14th Inverse Problems in Engineering Seminar, Ames, USA, (2006).
[13] F.J.C.P. Soeiro, P. Oliva Soares, A.J. Silva Neto, Solution of Inverse Radiative Transfer Problems with Artificial Neural Networks and Hybrid Methods, Proc. 13th Inverse Problems in Engineering Seminar, Cincinnatti, USA, (2004), 163–169.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish in this journal agree to the following terms:
Authors retain copyright and grant the journal the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License that allows the sharing of the work with acknowledgment of authorship and initial publication in this journal.
Authors are authorized to assume additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (eg, publish in an institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
Authors are allowed and encouraged to publish and distribute their work online (eg, in institutional repositories or on their personal page) at any point before or during the editorial process, as this can generate productive changes as well as increase impact and the citation of the published work (See The effect of open access).
This is an open access journal which means that all content is freely available without charge to the user or his/her institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles, or use them for any other lawful purpose, without asking prior permission from the publisher or the
author. This is in accordance with the BOAI definition of open access
Intellectual Property
All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License under attribution BY.