Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt
Autor(a) principal: | |
---|---|
Data de Publicação: | 2023 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFMG |
Texto Completo: | http://hdl.handle.net/1843/61122 |
Resumo: | Extracting geological resources like hydrocarbon fluids requires significant investments and precise decision-making processes. To optimize the efficiency of the extraction process, researchers and industry experts have explored innovative methodologies, including the prediction of optimal drilling locations. Porosity, a key attribute of reservoir rocks, plays a crucial role in determining fluid storage capacity. Geostatistical techniques, such as kriging, have been widely used for estimating porosity by capturing spatial dependence in sampled point-referenced data. However, the reliance on geographical coordinates for determining spatial distances may present challenges in scenarios with small and widely separated samples. In this paper, we develop a mixture model that combines the covariance generated by geographical space and the covariance generated in an appropriate feature space to enhance estimation accuracy. Developed within the Bayesian framework, our approach utilizes flexible Markov Chain Monte Carlo (MCMC) methods and leverages the Nearest-Neighbor Gaussian Process (NNGP) strategy for scalability. We present a controlled empirical comparison, considering various data generation configurations, to assess the performance of the mixture model in comparison to the marginal models. Applying our models to a three-dimensional reservoir simulation demonstrates its practical applicability and scalability. This research presents a novel approach for improved porosity estimation by integrating spatial and covariate information, offering the potential for optimizing reservoir exploration and extraction activities. |
id |
UFMG_20cb3e83dc77dd8d81de1271207d5c2e |
---|---|
oai_identifier_str |
oai:repositorio.ufmg.br:1843/61122 |
network_acronym_str |
UFMG |
network_name_str |
Repositório Institucional da UFMG |
repository_id_str |
|
spelling |
Marcos Oliveira Prateshttp://lattes.cnpq.br/7893235207392165Thais Paiva GallettiLuis Mauricio Castro Ceperohttp://lattes.cnpq.br/7938051491678752Lucas Azevedo Birro Michelin2023-11-20T11:48:57Z2023-11-20T11:48:57Z2023-08-15http://hdl.handle.net/1843/611220000-0001-8077-4898Extracting geological resources like hydrocarbon fluids requires significant investments and precise decision-making processes. To optimize the efficiency of the extraction process, researchers and industry experts have explored innovative methodologies, including the prediction of optimal drilling locations. Porosity, a key attribute of reservoir rocks, plays a crucial role in determining fluid storage capacity. Geostatistical techniques, such as kriging, have been widely used for estimating porosity by capturing spatial dependence in sampled point-referenced data. However, the reliance on geographical coordinates for determining spatial distances may present challenges in scenarios with small and widely separated samples. In this paper, we develop a mixture model that combines the covariance generated by geographical space and the covariance generated in an appropriate feature space to enhance estimation accuracy. Developed within the Bayesian framework, our approach utilizes flexible Markov Chain Monte Carlo (MCMC) methods and leverages the Nearest-Neighbor Gaussian Process (NNGP) strategy for scalability. We present a controlled empirical comparison, considering various data generation configurations, to assess the performance of the mixture model in comparison to the marginal models. Applying our models to a three-dimensional reservoir simulation demonstrates its practical applicability and scalability. This research presents a novel approach for improved porosity estimation by integrating spatial and covariate information, offering the potential for optimizing reservoir exploration and extraction activities.Extrair recursos geológicos, como fluidos de hidrocarbonetos, requer investimentos significativos e processos de tomada de decisão precisos. Para otimizar a eficiência do processo de extração, pesquisadores e especialistas da indústria têm explorado metodologias inovadoras, incluindo a previsão de locais de perfuração ótimos. A porosidade, um atributo fundamental das rochas de um reservatório, desempenha um papel crucial na determinação da sua capacidade de armazenamento de fluidos. Técnicas geoestatísticas, como a ''krigagem'', têm sido amplamente utilizadas para estimar a porosidade, capturando a dependência espacial em dados de amostras pontuais. No entanto, a dependência das coordenadas geográficas para determinar distâncias espaciais pode apresentar desafios em cenários de pequenas amostras e amplamente separadas. Neste artigo, desenvolvemos um modelo de mistura que combina a covariância gerada pelo espaço geográfico e a covariância gerada em um espaço de covariáveis (\textit{features}) apropriado para aprimorar a precisão da estimativa. Desenvolvido no contexto Bayesiano, nossa abordagem utiliza métodos de Monte Carlo com Cadeia de Markov (MCMC) e aproveita a estratégia do Processo Gaussiano dos vizinhos mais próximos (NNGP) para atingir escalabilidade. Apresentamos uma comparação em um estudo de simulação, considerando várias configurações para geração dos dados, a fim de avaliar o desempenho do modelo de mistura em comparação aos modelos marginais. Além disso, a aplicação dos nossos modelos em uma simulação de reservatório tridimensional demonstra sua aplicabilidade prática e escalabilidade. Esta pesquisa apresenta uma abordagem inovadora para a melhoria da estimativa de porosidade, integrando informações espaciais e de covariáveis, oferecendo o potencial para otimizar atividades de exploração e extração de reservatórios.engUniversidade Federal de Minas GeraisPrograma de Pós-Graduação em EstatísticaUFMGBrasilICX - DEPARTAMENTO DE ESTATÍSTICAEstatística – TesesEstatística espacial – TesesTeoria bayesiana de decisão estatística - TesesPetróleo - Pós sal- Brasil - TesesSpatial statisticsCokrigingComputation Bayesian MethodsFeature SpacePorosity EstimationFast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-saltinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFMGinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGORIGINALDissertação_Mestrado_2023_Lucas_Michelin_FinalVersion.pdfDissertação_Mestrado_2023_Lucas_Michelin_FinalVersion.pdfapplication/pdf6853514https://repositorio.ufmg.br/bitstream/1843/61122/1/Disserta%c3%a7%c3%a3o_Mestrado_2023_Lucas_Michelin_FinalVersion.pdfc1f87776b5cb7bfa994493f42d3b7194MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-82118https://repositorio.ufmg.br/bitstream/1843/61122/2/license.txtcda590c95a0b51b4d15f60c9642ca272MD521843/611222023-11-20 08:48:58.263oai:repositorio.ufmg.br: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ório de PublicaçõesPUBhttps://repositorio.ufmg.br/oaiopendoar:2023-11-20T11:48:58Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG)false |
dc.title.pt_BR.fl_str_mv |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
title |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
spellingShingle |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt Lucas Azevedo Birro Michelin Spatial statistics Cokriging Computation Bayesian Methods Feature Space Porosity Estimation Estatística – Teses Estatística espacial – Teses Teoria bayesiana de decisão estatística - Teses Petróleo - Pós sal- Brasil - Teses |
title_short |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
title_full |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
title_fullStr |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
title_full_unstemmed |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
title_sort |
Fast mixture spatial regression: a mixture in the geographical and feature space applied to predict oil in the post-salt |
author |
Lucas Azevedo Birro Michelin |
author_facet |
Lucas Azevedo Birro Michelin |
author_role |
author |
dc.contributor.advisor1.fl_str_mv |
Marcos Oliveira Prates |
dc.contributor.advisor1Lattes.fl_str_mv |
http://lattes.cnpq.br/7893235207392165 |
dc.contributor.referee1.fl_str_mv |
Thais Paiva Galletti |
dc.contributor.referee2.fl_str_mv |
Luis Mauricio Castro Cepero |
dc.contributor.authorLattes.fl_str_mv |
http://lattes.cnpq.br/7938051491678752 |
dc.contributor.author.fl_str_mv |
Lucas Azevedo Birro Michelin |
contributor_str_mv |
Marcos Oliveira Prates Thais Paiva Galletti Luis Mauricio Castro Cepero |
dc.subject.por.fl_str_mv |
Spatial statistics Cokriging Computation Bayesian Methods Feature Space Porosity Estimation |
topic |
Spatial statistics Cokriging Computation Bayesian Methods Feature Space Porosity Estimation Estatística – Teses Estatística espacial – Teses Teoria bayesiana de decisão estatística - Teses Petróleo - Pós sal- Brasil - Teses |
dc.subject.other.pt_BR.fl_str_mv |
Estatística – Teses Estatística espacial – Teses Teoria bayesiana de decisão estatística - Teses Petróleo - Pós sal- Brasil - Teses |
description |
Extracting geological resources like hydrocarbon fluids requires significant investments and precise decision-making processes. To optimize the efficiency of the extraction process, researchers and industry experts have explored innovative methodologies, including the prediction of optimal drilling locations. Porosity, a key attribute of reservoir rocks, plays a crucial role in determining fluid storage capacity. Geostatistical techniques, such as kriging, have been widely used for estimating porosity by capturing spatial dependence in sampled point-referenced data. However, the reliance on geographical coordinates for determining spatial distances may present challenges in scenarios with small and widely separated samples. In this paper, we develop a mixture model that combines the covariance generated by geographical space and the covariance generated in an appropriate feature space to enhance estimation accuracy. Developed within the Bayesian framework, our approach utilizes flexible Markov Chain Monte Carlo (MCMC) methods and leverages the Nearest-Neighbor Gaussian Process (NNGP) strategy for scalability. We present a controlled empirical comparison, considering various data generation configurations, to assess the performance of the mixture model in comparison to the marginal models. Applying our models to a three-dimensional reservoir simulation demonstrates its practical applicability and scalability. This research presents a novel approach for improved porosity estimation by integrating spatial and covariate information, offering the potential for optimizing reservoir exploration and extraction activities. |
publishDate |
2023 |
dc.date.accessioned.fl_str_mv |
2023-11-20T11:48:57Z |
dc.date.available.fl_str_mv |
2023-11-20T11:48:57Z |
dc.date.issued.fl_str_mv |
2023-08-15 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/1843/61122 |
dc.identifier.orcid.pt_BR.fl_str_mv |
0000-0001-8077-4898 |
url |
http://hdl.handle.net/1843/61122 |
identifier_str_mv |
0000-0001-8077-4898 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.publisher.program.fl_str_mv |
Programa de Pós-Graduação em Estatística |
dc.publisher.initials.fl_str_mv |
UFMG |
dc.publisher.country.fl_str_mv |
Brasil |
dc.publisher.department.fl_str_mv |
ICX - DEPARTAMENTO DE ESTATÍSTICA |
publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFMG instname:Universidade Federal de Minas Gerais (UFMG) instacron:UFMG |
instname_str |
Universidade Federal de Minas Gerais (UFMG) |
instacron_str |
UFMG |
institution |
UFMG |
reponame_str |
Repositório Institucional da UFMG |
collection |
Repositório Institucional da UFMG |
bitstream.url.fl_str_mv |
https://repositorio.ufmg.br/bitstream/1843/61122/1/Disserta%c3%a7%c3%a3o_Mestrado_2023_Lucas_Michelin_FinalVersion.pdf https://repositorio.ufmg.br/bitstream/1843/61122/2/license.txt |
bitstream.checksum.fl_str_mv |
c1f87776b5cb7bfa994493f42d3b7194 cda590c95a0b51b4d15f60c9642ca272 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG) |
repository.mail.fl_str_mv |
|
_version_ |
1803589157140824064 |