Drilling optimization of petroleum and natural gas wells: application of artificial intelligence
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://hdl.handle.net/11449/180784 |
Resumo: | To meet the increasing primary energy demand, more challenging petroleum reserves started being explored, such as the reservoirs from pre-salt formation close to the Brazilian and Angolan coasts. Historically, low penetration rates in drilling the pre-salt carbonates were reported in the literature, resulting in large capital expenditure on well’s construction. Since the major part of exploration cost is associated with drilling, optimizing this activity is of major importance. In this context, the main objective of the present thesis is to investigate methods for real-time drilling optimization of oil and natural gas wells. A common way to optimize drilling activities is to determine the optimum operational variables (e.g. weight-on-bit and rotational speed) that maximizes the ROP. However, this may yield a decrease in drilling efficiency. An alternative to reduce problems related to drilling inefficiency, such as excessive bit wear and vibrations, is through the selection of operational variables able to minimize the specific energy (SE) spent to excavate a volumetric unit of rock. For that, it is necessary to employ accurate predictive models able to capture how the operational variables (weight-on-bit, rotational speed, mud flow and so on) influence not only on ROP but also on SE. Therefore, the present thesis employed a well-known machine learning method, called random forest, instead of analytical equations found in drilling engineering books. Thus, it was possible to obtain accurate predictive models for ROP and SE, to be used, later, as objective functions in optimization problems to determine the optimum parameters, weight-on-bit and rotational speed. Real-time drilling data from pre-salt region and Norwegian continental shelf were employed. First, several aspects related to training process of random forests were investigated. Among them, it was confirmed the possibility of predicting the ROP with accuracy by employing only four inputs: depth, weight-on-bit, rotational speed, and mud flow. The prediction of SE was carried out by coupling the mathematical formulation with predictive models of ROP and torque (if available). Optimization problems were analyzed with one objective function, as well as with multiple objective functions through the ε-constraint technique. It was observed the sole maximization of ROP may lead to increase in the energy required to drill. However, by imposing the inequality SE(x) ≤ SEacutal*ε on the maximization of ROP, it was possible to reduce significantly the amount of observations whose ROP increased due to detriment of drilling efficiency. For the minimization of SE problems, it was observed a special care to be taken when simulating low-values for weigh-on-bit and rotational speed. |
id |
UNSP_5ca1f2b47d3f2ecb1b4464b5c9d07127 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/180784 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligenceOtimização da perfuração de poços de petróleos e gás naturais: aplicação da inteligência artificialMulti-objective optimizationDrilling efficiencyRate of penetrationMachine learningRandom forestOtimização multiobjetivaEficiência da perfuraçãoTaxa de penetraçãoAprendizagem de máquinasFlorestas aleatóriasMáquinas de perfuraçãoPoços de petróleo - PerfuraçãoGás naturalIndústria petrolíferaTo meet the increasing primary energy demand, more challenging petroleum reserves started being explored, such as the reservoirs from pre-salt formation close to the Brazilian and Angolan coasts. Historically, low penetration rates in drilling the pre-salt carbonates were reported in the literature, resulting in large capital expenditure on well’s construction. Since the major part of exploration cost is associated with drilling, optimizing this activity is of major importance. In this context, the main objective of the present thesis is to investigate methods for real-time drilling optimization of oil and natural gas wells. A common way to optimize drilling activities is to determine the optimum operational variables (e.g. weight-on-bit and rotational speed) that maximizes the ROP. However, this may yield a decrease in drilling efficiency. An alternative to reduce problems related to drilling inefficiency, such as excessive bit wear and vibrations, is through the selection of operational variables able to minimize the specific energy (SE) spent to excavate a volumetric unit of rock. For that, it is necessary to employ accurate predictive models able to capture how the operational variables (weight-on-bit, rotational speed, mud flow and so on) influence not only on ROP but also on SE. Therefore, the present thesis employed a well-known machine learning method, called random forest, instead of analytical equations found in drilling engineering books. Thus, it was possible to obtain accurate predictive models for ROP and SE, to be used, later, as objective functions in optimization problems to determine the optimum parameters, weight-on-bit and rotational speed. Real-time drilling data from pre-salt region and Norwegian continental shelf were employed. First, several aspects related to training process of random forests were investigated. Among them, it was confirmed the possibility of predicting the ROP with accuracy by employing only four inputs: depth, weight-on-bit, rotational speed, and mud flow. The prediction of SE was carried out by coupling the mathematical formulation with predictive models of ROP and torque (if available). Optimization problems were analyzed with one objective function, as well as with multiple objective functions through the ε-constraint technique. It was observed the sole maximization of ROP may lead to increase in the energy required to drill. However, by imposing the inequality SE(x) ≤ SEacutal*ε on the maximization of ROP, it was possible to reduce significantly the amount of observations whose ROP increased due to detriment of drilling efficiency. For the minimization of SE problems, it was observed a special care to be taken when simulating low-values for weigh-on-bit and rotational speed.Para atender à crescente demanda de energia primária, começaram a ser exploradas reservas de petróleo em áreas mais desafiadoras, tais como os reservatórios da formação do pré-sal próximos às costas brasileira e angolana. Historicamente, observa-se baixa taxa de penetração na perfuração dos carbonatos do pré-sal, resultando em altos custos na construção de poços. Como a maior parte dos custos de exploração está associado com perfuração, a otimização desta atividade é de grande importância. Neste contexto, o principal objetivo desta dissertação é investigar métodos de otimização em tempo-real de poços de petróleo e gás natural. Uma forma comum de se otimizar a perfuração é através da determinação dos parâmetros operacionais (peso na broca e rotação) que maximizem a taxa de penetração (rate of penetration, ROP). Contudo, isto pode acarretar na diminuição da eficiência do processo de perfuração. Assim, uma forma de diminuir problemas relacionadas a ineficiências da perfuração, tais como gasto excessivo da broca ou vibrações, é através da seleção dos parâmetros operacionais, minimizando a energia específica (specific energy, SE) gasta para escavar uma unidade volumétrica de rocha. Para tanto, é necessário o emprego de modelos precisos que relacionem como as variáveis operacionais (peso da broca, rotação, vazão do fluido de perfuração entre outros) influenciam, não somente o ROP, mas também a SE. Desde modo, a presente dissertação empregou um método conhecido de aprendizagem de máquinas, chamado de florestas aleatórias, em vez das equações analíticas comumente encontrados em livros de engenharia de perfuração. Assim, foi possível obter modelos de previsão precisos para ROP e SE, para, depois, serem utilizados como funções objetivas em problemas de otimização para seleção ótima dos parâmetros (peso na broca e rotação). Dados de perfuração da região do pré-sal e da plataforma continental norueguesa foram utilizados. Primeiramente, investigou-se diversos aspectos relacionados ao treinamento das florestas aleatórias. Entre eles, verificou-se a possibilidade de estimar com precisão o ROP utilizando apenas quatro parâmetros: profundidade, peso na broca, rotação e vazão de fluido. A previsão da SE, por sua vez, se fez através do acoplamento da formulação matemática com os modelos preditivos do ROP e torque (quando disponível). Foram investigados problemas de otimização contendo tanto uma função objetiva quanto problemas com múltiplas funções objetivas através da técnica "ε-constraint". Verificou-se que a maximização sozinha da taxa de penetração pode acarretar em aumento da energia gasta para se perfurar. Contudo, ao impor como restrição a inequação SE(x) ≤ SEactual*ε ao problema da maximização da taxa de penetração, foi possível diminuir consideravelmente a quantidade de observações que o aumento do ROP se deu através do detrimento da eficiência da perfuração. Para o problema da minimização da SE, constatou um cuidado que se deve ter ao simular combinações de peso na broca e rotação com valores baixos.Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)001Universidade Estadual Paulista (Unesp)Carvalho Junior, João Andrade de [UNESP]Nascimento, Andreas [UNESP]Universidade Estadual Paulista (Unesp)Barbosa, Luís Felipe Ferreira Motta2019-02-18T14:18:48Z2019-02-18T14:18:48Z2019-01-11info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/11449/18078400091285033004080027P6enginfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESP2023-10-12T06:06:56Zoai:repositorio.unesp.br:11449/180784Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462023-10-12T06:06:56Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence Otimização da perfuração de poços de petróleos e gás naturais: aplicação da inteligência artificial |
title |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
spellingShingle |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence Barbosa, Luís Felipe Ferreira Motta Multi-objective optimization Drilling efficiency Rate of penetration Machine learning Random forest Otimização multiobjetiva Eficiência da perfuração Taxa de penetração Aprendizagem de máquinas Florestas aleatórias Máquinas de perfuração Poços de petróleo - Perfuração Gás natural Indústria petrolífera |
title_short |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
title_full |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
title_fullStr |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
title_full_unstemmed |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
title_sort |
Drilling optimization of petroleum and natural gas wells: application of artificial intelligence |
author |
Barbosa, Luís Felipe Ferreira Motta |
author_facet |
Barbosa, Luís Felipe Ferreira Motta |
author_role |
author |
dc.contributor.none.fl_str_mv |
Carvalho Junior, João Andrade de [UNESP] Nascimento, Andreas [UNESP] Universidade Estadual Paulista (Unesp) |
dc.contributor.author.fl_str_mv |
Barbosa, Luís Felipe Ferreira Motta |
dc.subject.por.fl_str_mv |
Multi-objective optimization Drilling efficiency Rate of penetration Machine learning Random forest Otimização multiobjetiva Eficiência da perfuração Taxa de penetração Aprendizagem de máquinas Florestas aleatórias Máquinas de perfuração Poços de petróleo - Perfuração Gás natural Indústria petrolífera |
topic |
Multi-objective optimization Drilling efficiency Rate of penetration Machine learning Random forest Otimização multiobjetiva Eficiência da perfuração Taxa de penetração Aprendizagem de máquinas Florestas aleatórias Máquinas de perfuração Poços de petróleo - Perfuração Gás natural Indústria petrolífera |
description |
To meet the increasing primary energy demand, more challenging petroleum reserves started being explored, such as the reservoirs from pre-salt formation close to the Brazilian and Angolan coasts. Historically, low penetration rates in drilling the pre-salt carbonates were reported in the literature, resulting in large capital expenditure on well’s construction. Since the major part of exploration cost is associated with drilling, optimizing this activity is of major importance. In this context, the main objective of the present thesis is to investigate methods for real-time drilling optimization of oil and natural gas wells. A common way to optimize drilling activities is to determine the optimum operational variables (e.g. weight-on-bit and rotational speed) that maximizes the ROP. However, this may yield a decrease in drilling efficiency. An alternative to reduce problems related to drilling inefficiency, such as excessive bit wear and vibrations, is through the selection of operational variables able to minimize the specific energy (SE) spent to excavate a volumetric unit of rock. For that, it is necessary to employ accurate predictive models able to capture how the operational variables (weight-on-bit, rotational speed, mud flow and so on) influence not only on ROP but also on SE. Therefore, the present thesis employed a well-known machine learning method, called random forest, instead of analytical equations found in drilling engineering books. Thus, it was possible to obtain accurate predictive models for ROP and SE, to be used, later, as objective functions in optimization problems to determine the optimum parameters, weight-on-bit and rotational speed. Real-time drilling data from pre-salt region and Norwegian continental shelf were employed. First, several aspects related to training process of random forests were investigated. Among them, it was confirmed the possibility of predicting the ROP with accuracy by employing only four inputs: depth, weight-on-bit, rotational speed, and mud flow. The prediction of SE was carried out by coupling the mathematical formulation with predictive models of ROP and torque (if available). Optimization problems were analyzed with one objective function, as well as with multiple objective functions through the ε-constraint technique. It was observed the sole maximization of ROP may lead to increase in the energy required to drill. However, by imposing the inequality SE(x) ≤ SEacutal*ε on the maximization of ROP, it was possible to reduce significantly the amount of observations whose ROP increased due to detriment of drilling efficiency. For the minimization of SE problems, it was observed a special care to be taken when simulating low-values for weigh-on-bit and rotational speed. |
publishDate |
2019 |
dc.date.none.fl_str_mv |
2019-02-18T14:18:48Z 2019-02-18T14:18:48Z 2019-01-11 |
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/11449/180784 000912850 33004080027P6 |
url |
http://hdl.handle.net/11449/180784 |
identifier_str_mv |
000912850 33004080027P6 |
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.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) |
publisher.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799964533302755328 |