Documents disponibles écrits par cet auteur

Using artificial neural networks to generate synthetic well logs / Luisa Rolon in Journal of natural gas science and engineering, Vol. 1 N° 4-5 (Novembre 2009) 

Public ISBD [article]  in Journal of natural gas science and engineering > Vol. 1 N° 4-5 (Novembre 2009) . - pp. 118-133 Titre : Using artificial neural networks to generate synthetic well logs Type de document : texte imprimé Auteurs : Luisa Rolon, Auteur ; Shahab D. Mohaghegh, Auteur ; Sam Ameri, Auteur Article en page(s) : pp. 118-133 Note générale : Génie Chimique Langues : Anglais (eng) Mots-clés : Synthetic  logs  Neural  networks  Reservoir  characterization  Well  logs  Artificial  intelligence  Petrophysics Index. décimale : 665.7 Résumé : A methodology to generate synthetic wireline logs is presented. Synthetic logs can help to analyze the reservoir properties in areas where the set of logs that are necessary, are absent or incomplete. The approach presented involves the use of artificial neural networks as the main tool, in conjunction with data obtained from conventional wireline logs. Implementation of this approach aims to reduce costs to companies. Development of the neural network model was completed using Generalized Regression Neural Network, and wireline logs from four wells that included gamma ray, density, neutron, and resistivity logs. Synthetic logs were generated through two different exercises. Exercise one involved all four wells for training, calibration and verification process. The second exercise used three wells for training and calibration and the fourth well was used for verification. In order to demonstrate the robustness of the methodology, three different combinations of inputs/outputs were chosen to train the network. In combination “A” the resistivity log was the output and density, gamma ray, and neutron logs, and the coordinates and depths (XYZ) the inputs. In combination “B” the density log was output and the resistivity, the gamma ray, and the neutron logs, and XYZ were the inputs, and in combination “C” the neutron log was the output while the resistivity, the gamma ray, and the density logs, and XYZ were the inputs. After development of the neural network model, synthetic logs with a reasonable degree of accuracy were generated. Results indicate that the best performance was obtained for combination “A” of inputs and outputs, then for combination “C”, and finally for combination “B”. In addition, it was determined that accuracy of synthetic logs is favored by interpolation of data. It was also demonstrated that using neural network to generate synthetic well logs is far more superior when compared to conventional approaches such as multiple-regression. DEWEY : 665.7 ISSN : 1875-5100 En ligne : http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%2356453%2 [...] [article] Using artificial neural networks to generate synthetic well logs [texte imprimé] / Luisa Rolon, Auteur ; Shahab D. Mohaghegh, Auteur ; Sam Ameri, Auteur . - pp. 118-133. Génie Chimique Langues : Anglais (eng) in Journal of natural gas science and engineering > Vol. 1 N° 4-5 (Novembre 2009) . - pp. 118-133 Mots-clés : Synthetic  logs  Neural  networks  Reservoir  characterization  Well  logs  Artificial  intelligence  Petrophysics Index. décimale : 665.7 Résumé : A methodology to generate synthetic wireline logs is presented. Synthetic logs can help to analyze the reservoir properties in areas where the set of logs that are necessary, are absent or incomplete. The approach presented involves the use of artificial neural networks as the main tool, in conjunction with data obtained from conventional wireline logs. Implementation of this approach aims to reduce costs to companies. Development of the neural network model was completed using Generalized Regression Neural Network, and wireline logs from four wells that included gamma ray, density, neutron, and resistivity logs. Synthetic logs were generated through two different exercises. Exercise one involved all four wells for training, calibration and verification process. The second exercise used three wells for training and calibration and the fourth well was used for verification. In order to demonstrate the robustness of the methodology, three different combinations of inputs/outputs were chosen to train the network. In combination “A” the resistivity log was the output and density, gamma ray, and neutron logs, and the coordinates and depths (XYZ) the inputs. In combination “B” the density log was output and the resistivity, the gamma ray, and the neutron logs, and XYZ were the inputs, and in combination “C” the neutron log was the output while the resistivity, the gamma ray, and the density logs, and XYZ were the inputs. After development of the neural network model, synthetic logs with a reasonable degree of accuracy were generated. Results indicate that the best performance was obtained for combination “A” of inputs and outputs, then for combination “C”, and finally for combination “B”. In addition, it was determined that accuracy of synthetic logs is favored by interpolation of data. It was also demonstrated that using neural network to generate synthetic well logs is far more superior when compared to conventional approaches such as multiple-regression. DEWEY : 665.7 ISSN : 1875-5100 En ligne : http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%2356453%2 [...]