L. Hoàng, Minh Quý Nguyễn, Vũ Anh Phan, Thị Thu Hường Lê, Thế Hùng Lê, Linh Hoàng, Việt Dũng Bùi, Văn Độ Nguyễn
Tình trạng nhiễm bẩn vùng cận đáy giếng, thiết bị lòng giếng và trong giếng khai thác có thể do dung dịch khoan gây ra trong quá trình khoan mở vỉa tầng sản phẩm; hoặc do trong quá trình khai thác xuất hiện hiện tượng cát xâm nhập, độ ngập nước tăng cao, lắng đọng paraffin, asphaltene, lắng đọng cặn sa lắng vô cơ; do sự thay đổi lớn và đột ngột về các thông số động học như áp suất và nhiệt độ tại vùng cận đáy giếng làm thay đổi tính chất lý hóa, phá vỡ trạng thái cân bằng pha của các lưu thể, hoặc quá trình tạo nhũ tương, thay đổi tính dính ướt và mối quan hệ dòng chảy.Nghiên cứu đã đánh giá hiện trạng hoạt động của các giếng khai thác để xác định nguyên nhân chính gây ra tình trạng nhiễm bẩn vùng cận đáy giếng của các giếng ở bể Cửu Long. Trên cơ sở xác định được cơ chế nhiễm bẩn chính là do quá trình hình thành các muối vô cơ với phần nhỏ kết dính của cặn hữu cơ và quá trình dịch chuyển các khoáng vật sét, hạt mịn gây bít nhét, cản trở dòng chảy của chất lưu khai thác tại các mỏ, nhóm tác giả đề xuất giải pháp tối ưu xử lý vùng cận đáy giếng cho các giếng khai thác dầu tại bể Cửu Long và bể Nam Côn Sơn. Các giải pháp xử lý acid tối ưu cho vùng cận đáy giếng sẽ góp phần giảm thiểu rủi ro, nâng cao hiệu quả khai thác và phục vụ công tác quản lý, điều hành mỏ.
{"title":"Nguyên nhân nhiễm bẩn vùng cận đáy giếng và giải pháp xử lý cho các giếng khai thác tại bể Cửu Long và bể Nam Côn Sơn, thềm lục địa Việt Nam","authors":"L. Hoàng, Minh Quý Nguyễn, Vũ Anh Phan, Thị Thu Hường Lê, Thế Hùng Lê, Linh Hoàng, Việt Dũng Bùi, Văn Độ Nguyễn","doi":"10.47800/pvj.2022.07-01","DOIUrl":"https://doi.org/10.47800/pvj.2022.07-01","url":null,"abstract":"Tình trạng nhiễm bẩn vùng cận đáy giếng, thiết bị lòng giếng và trong giếng khai thác có thể do dung dịch khoan gây ra trong quá trình khoan mở vỉa tầng sản phẩm; hoặc do trong quá trình khai thác xuất hiện hiện tượng cát xâm nhập, độ ngập nước tăng cao, lắng đọng paraffin, asphaltene, lắng đọng cặn sa lắng vô cơ; do sự thay đổi lớn và đột ngột về các thông số động học như áp suất và nhiệt độ tại vùng cận đáy giếng làm thay đổi tính chất lý hóa, phá vỡ trạng thái cân bằng pha của các lưu thể, hoặc quá trình tạo nhũ tương, thay đổi tính dính ướt và mối quan hệ dòng chảy.Nghiên cứu đã đánh giá hiện trạng hoạt động của các giếng khai thác để xác định nguyên nhân chính gây ra tình trạng nhiễm bẩn vùng cận đáy giếng của các giếng ở bể Cửu Long. Trên cơ sở xác định được cơ chế nhiễm bẩn chính là do quá trình hình thành các muối vô cơ với phần nhỏ kết dính của cặn hữu cơ và quá trình dịch chuyển các khoáng vật sét, hạt mịn gây bít nhét, cản trở dòng chảy của chất lưu khai thác tại các mỏ, nhóm tác giả đề xuất giải pháp tối ưu xử lý vùng cận đáy giếng cho các giếng khai thác dầu tại bể Cửu Long và bể Nam Côn Sơn. Các giải pháp xử lý acid tối ưu cho vùng cận đáy giếng sẽ góp phần giảm thiểu rủi ro, nâng cao hiệu quả khai thác và phục vụ công tác quản lý, điều hành mỏ.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126628601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thanh Phu Nguyen, Van Hoanh Nguyen, Q. Ta, The Ha Le
Gas temperature is an essential parameter in estimating production rate and pressure model inside the production tubing. Three heat transfer mechanisms named as conduction, convection and radiation have been applied to identify the gas temperature declination. Gas wells with bottom hole temperature greater than 160oC and gas rates reaching 55 million standard ft3 per day (MMscf/d) indicate a higher heat loss due to convection than the other two mechanisms. Conduction is the main factor in explaining heat diffusion to the surrounding at the top of the well. The study presents a strong similarity in value compared to the field data by combining Gray correlation and heat transfer model to predict the bottom hole pressure with an error of approximately 3%. Additionally, the gas temperature affects gas rate prediction through gas viscosity and Z factor. With the gas composition mostly containing C1 (70.5%), gas viscosity and Z coefficient at the wellhead are not as high as 0.017 cp and 0.92 respectively. It is possible to have a two-phase flow, then a temperature model along the production tubing is necessary to ensure the gas production rate.
{"title":"Assessing the effect of gas temperature on gas well performance","authors":"Thanh Phu Nguyen, Van Hoanh Nguyen, Q. Ta, The Ha Le","doi":"10.47800/pvj.2022.06-06","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-06","url":null,"abstract":"Gas temperature is an essential parameter in estimating production rate and pressure model inside the production tubing. Three heat transfer mechanisms named as conduction, convection and radiation have been applied to identify the gas temperature declination. Gas wells with bottom hole temperature greater than 160oC and gas rates reaching 55 million standard ft3 per day (MMscf/d) indicate a higher heat loss due to convection than the other two mechanisms. Conduction is the main factor in explaining heat diffusion to the surrounding at the top of the well. The study presents a strong similarity in value compared to the field data by combining Gray correlation and heat transfer model to predict the bottom hole pressure with an error of approximately 3%. Additionally, the gas temperature affects gas rate prediction through gas viscosity and Z factor. With the gas composition mostly containing C1 (70.5%), gas viscosity and Z coefficient at the wellhead are not as high as 0.017 cp and 0.92 respectively. It is possible to have a two-phase flow, then a temperature model along the production tubing is necessary to ensure the gas production rate.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134264296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The purpose of the study is to activate a well named X (for confidential reasons) in order to improve its production by proposing an electrical submersible pump. The nodal analysis is performed to understand the well’s condition and an economic evaluation is done to determine the applicability of the project. The initial completion data, the pump placement data and the economic data are considered and used as input in PIPESIM 2017 software for operations and simulations. The results obtained from nodal analysis show that the well is in a total depletion situation. Upon analysis, the electrical submersible pump type REDA S6000N with operational diameter of 5.38 inches is appropriately chosen and installed, resulting in a flowrate of 4,891.36 stock-tank barrels per day (stb/d) with a bottom pressure of 2,735 pounds per square inch (psi). A flowrate of 5,000 stock-tank barrels per day at a pressure of 2,707 psi is obtained after optimisation of the pump through sensitivity curves. The economic balance sheet presents a net present value of USD 110,718,250, showing the profitability of the project over a period of one year.
{"title":"Activation of a non-eruptive well by using an electrical pump to optimise production","authors":"Belomo Victorine, Nitcheu Madeleine, Dongmo Eric Donald, Njeudjang Kasi, Kuiatse Gabriel, Takougang Kingni Sifeu","doi":"10.47800/pvj.2022.06-04","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-04","url":null,"abstract":"The purpose of the study is to activate a well named X (for confidential reasons) in order to improve its production by proposing an electrical submersible pump. The nodal analysis is performed to understand the well’s condition and an economic evaluation is done to determine the applicability of the project. The initial completion data, the pump placement data and the economic data are considered and used as input in PIPESIM 2017 software for operations and simulations. The results obtained from nodal analysis show that the well is in a total depletion situation. Upon analysis, the electrical submersible pump type REDA S6000N with operational diameter of 5.38 inches is appropriately chosen and installed, resulting in a flowrate of 4,891.36 stock-tank barrels per day (stb/d) with a bottom pressure of 2,735 pounds per square inch (psi). A flowrate of 5,000 stock-tank barrels per day at a pressure of 2,707 psi is obtained after optimisation of the pump through sensitivity curves. The economic balance sheet presents a net present value of USD 110,718,250, showing the profitability of the project over a period of one year.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124684763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Matateyou, K. Njeudjang, Eric Donald Dongmo, Cherif Ibrahim Rengou Mbouombouo, Hugues Vannela Kamnang Konchipe, Gabriel Kuiatse
The present paper aims to design a continuous gas-lift system in order to activate the dead well X and optimise the recovery of hydrocarbons by gas injection. The data taken into account are those of the reservoir and the well. All the simulation operations and the well diagram are carried out with the PIPESIM 2017 2.0 software. Moreover, the nodal analysis of the dead well X is carried out by finding the injection pressure of the gas in the well. The number of valves to be installed, the optimal injection heights and the flow rates received by each valve are also included in the gas-lift device. Then the evaluation of the system performance as well as the sensitivity analysis are carried out to make it possible to fix the optimal flow rate of production. The results obtained show that for a continuous gas injection of 3 million standard ft3 per day (MMscf/d) per valve into the well with a wellhead pressure of 2,500 psi and an optimal flow rate of 2,718 standard barrels per day (STB/d), a profitability of USD 182,887,219 is obtained. In this field, this design can be applied to all wells having a production tube of 2.5 inches inside diameter. On the other hand, wells with different diameters will require a new evaluation.
{"title":"Design of continuous gas-lift for a dead well and step-up of its productivity","authors":"J. Matateyou, K. Njeudjang, Eric Donald Dongmo, Cherif Ibrahim Rengou Mbouombouo, Hugues Vannela Kamnang Konchipe, Gabriel Kuiatse","doi":"10.47800/pvj.2022.06-05","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-05","url":null,"abstract":"The present paper aims to design a continuous gas-lift system in order to activate the dead well X and optimise the recovery of hydrocarbons by gas injection. The data taken into account are those of the reservoir and the well. All the simulation operations and the well diagram are carried out with the PIPESIM 2017 2.0 software. Moreover, the nodal analysis of the dead well X is carried out by finding the injection pressure of the gas in the well. The number of valves to be installed, the optimal injection heights and the flow rates received by each valve are also included in the gas-lift device. Then the evaluation of the system performance as well as the sensitivity analysis are carried out to make it possible to fix the optimal flow rate of production. The results obtained show that for a continuous gas injection of 3 million standard ft3 per day (MMscf/d) per valve into the well with a wellhead pressure of 2,500 psi and an optimal flow rate of 2,718 standard barrels per day (STB/d), a profitability of USD 182,887,219 is obtained. In this field, this design can be applied to all wells having a production tube of 2.5 inches inside diameter. On the other hand, wells with different diameters will require a new evaluation.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121034903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. T. Kieu, Q. Pham, Quang Man Ha, H. G. Phạm, H. Doan, V. Bui, Hong Trang Pham
Petrophysical properties such as porosity, permeability and fluid saturations are important parameters for reservoir characterisation, which can be determined by experimental constitutive equations between rock parameters and well logging data. Thus, the same rock properties might demonstrate different patterns, depending on the input and equations used. In this work, we used the cross-properties (a common set of rock properties) that influence different measurements to reduce the ambiguity of the petrophysical property definition. We present an approach of using fuzzy c-means clustering to classify the well logs and core data in clusters and then running inversion for each cluster. The obtained results allowed us to establish suitable parameters in constitutive equations, which usually vary with rock units that may relate to clusters. Testing data applied to the Nam Con Son basin show a square correlation coefficient of 0.66 between the predicted and core measurement, suggesting a reasonable matching of the testing data set.
{"title":"Porosity prediction using fuzzy clustering and joint inversion of wireline logs: A case study of the Nam Con Son basin, offshore Vietnam","authors":"D. T. Kieu, Q. Pham, Quang Man Ha, H. G. Phạm, H. Doan, V. Bui, Hong Trang Pham","doi":"10.47800/pvj.2022.06-01","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-01","url":null,"abstract":"Petrophysical properties such as porosity, permeability and fluid saturations are important parameters for reservoir characterisation, which can be determined by experimental constitutive equations between rock parameters and well logging data. Thus, the same rock properties might demonstrate different patterns, depending on the input and equations used. In this work, we used the cross-properties (a common set of rock properties) that influence different measurements to reduce the ambiguity of the petrophysical property definition. We present an approach of using fuzzy c-means clustering to classify the well logs and core data in clusters and then running inversion for each cluster. The obtained results allowed us to establish suitable parameters in constitutive equations, which usually vary with rock units that may relate to clusters. Testing data applied to the Nam Con Son basin show a square correlation coefficient of 0.66 between the predicted and core measurement, suggesting a reasonable matching of the testing data set.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129691743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sandstone reservoirs are major reservoirs in siliciclastic rocks worldwide. A good understanding of the development of internal rock properties is, therefore, extremely important, especially in terms of porosity and permeability (which indicate reservoir storage and flow capacity), which are controlled by mineral compositions, rock textures, and diagenetic processes. This paper studied formations E and F in three wells in the Cuu Long basin to better define the impacts of not only depositional characters but also diagenetic overprints on porosity and permeability (poroperm). Core samples were analysed via thin section observations, scanning electron microscopy (SEM), X-ray diffraction (XRD) observations, capillary pressure (Pc) and helium porosity - permeability measurements together with petrophysical evaluation. Formation E was deposited in a fluvial - lacustrine environment that is characterised by claystone/shale interbedded with sandstone, with reduced depositional permeability in finer-grained intervals. XRD and SEM analyses indicate rock quality in the sandstone reservoirs was influenced by a variety of authigenic minerals, such as carbonate cements, quartz overgrowths, zeolites, and laumontite clays, which all tend to reduce poroperm. Whereas, formation F was deposited in a higher energy setting. This was mostly a braided channel environment indicated by a blocky shape in the wireline across the sandy interval and typically good primary porosity and permeability. In formation F, the reservoir quality is strongly controlled by diagenetic evolution. Pore throats in the E and F sandstones are reduced in size by intense compaction and a combination of pore-filling minerals including calcite cements, authigenic clays, and quartz overgrowths, leading to a negative relationship with poroperm. However, this negative relationship is not as clear in the formation E.
{"title":"Impact of deposition and diagenesis on quality of sandstone reservoirs: A case study in Cuu Long basin, offshore Vietnam","authors":"T. S. Nguyen","doi":"10.47800/pvj.2022.06-02","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-02","url":null,"abstract":"Sandstone reservoirs are major reservoirs in siliciclastic rocks worldwide. A good understanding of the development of internal rock properties is, therefore, extremely important, especially in terms of porosity and permeability (which indicate reservoir storage and flow capacity), which are controlled by mineral compositions, rock textures, and diagenetic processes. This paper studied formations E and F in three wells in the Cuu Long basin to better define the impacts of not only depositional characters but also diagenetic overprints on porosity and permeability (poroperm). Core samples were analysed via thin section observations, scanning electron microscopy (SEM), X-ray diffraction (XRD) observations, capillary pressure (Pc) and helium porosity - permeability measurements together with petrophysical evaluation. \u0000Formation E was deposited in a fluvial - lacustrine environment that is characterised by claystone/shale interbedded with sandstone, with reduced depositional permeability in finer-grained intervals. XRD and SEM analyses indicate rock quality in the sandstone reservoirs was influenced by a variety of authigenic minerals, such as carbonate cements, quartz overgrowths, zeolites, and laumontite clays, which all tend to reduce poroperm. Whereas, formation F was deposited in a higher energy setting. This was mostly a braided channel environment indicated by a blocky shape in the wireline across the sandy interval and typically good primary porosity and permeability. In formation F, the reservoir quality is strongly controlled by diagenetic evolution. Pore throats in the E and F sandstones are reduced in size by intense compaction and a combination of pore-filling minerals including calcite cements, authigenic clays, and quartz overgrowths, leading to a negative relationship with poroperm. However, this negative relationship is not as clear in the formation E.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127198598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thi Phuong Nhung Nguyen, T. Nguyen, H. L. Nguyen, T. Tran, V. Nguyen, P. Nguyen
This paper presents a simple process of fabricating a superhydrophobic carbon steel surface with high water repellence and good corrosion resistance. The steel surface was covered by thin ZnO layers via various methods of sol-gel method, hydrothermal synthesis, and electrochemical deposition, then further functionalised with methyltrichlorosilane. Morphological features observed by a scanning electron microscope (SEM) show that ZnO layers are composed of micro/nanostructures. Furthermore, the chemical composition of these films has been analysed by an energy dispersive X-ray spectroscopy (EDX). An optimum experimental condition, based on the type of Zn deposition method, has been obtained to provide superhydrophobic properties with a water contact angle of 152 ± 2o. In a 3.5 wt% NaCl solution, the corrosion resistance properties of the superhydrophobic ZnO-coated steel are found to be very good with an anti-corrosion efficiency of 93.12%.
{"title":"Micro/nanostructured ZnO-based superhydrophobic steel surface with enhanced corrosion protection","authors":"Thi Phuong Nhung Nguyen, T. Nguyen, H. L. Nguyen, T. Tran, V. Nguyen, P. Nguyen","doi":"10.47800/pvj.2022.06-07","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-07","url":null,"abstract":"This paper presents a simple process of fabricating a superhydrophobic carbon steel surface with high water repellence and good corrosion resistance. The steel surface was covered by thin ZnO layers via various methods of sol-gel method, hydrothermal synthesis, and electrochemical deposition, then further functionalised with methyltrichlorosilane. Morphological features observed by a scanning electron microscope (SEM) show that ZnO layers are composed of micro/nanostructures. Furthermore, the chemical composition of these films has been analysed by an energy dispersive X-ray spectroscopy (EDX). An optimum experimental condition, based on the type of Zn deposition method, has been obtained to provide superhydrophobic properties with a water contact angle of 152 ± 2o. In a 3.5 wt% NaCl solution, the corrosion resistance properties of the superhydrophobic ZnO-coated steel are found to be very good with an anti-corrosion efficiency of 93.12%.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"48 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115784590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
N. Nguyen, Ngoc The Hung Tran, Ky Son Hoang, Vũ Tùng Trần
Conventional integration of rock physics and seismic inversion can quantitatively evaluate and contrast reservoir properties. However, the available output attributes are occasionally not a perfect indicator for specific information such as lithology or fluid saturation due to technology constraints. Each attribute commonly exhibits a combination of geological characteristics that could lead to subjective interpretations and provides only qualitative results. Meanwhile, machine learning (ML) is emerging as an independent interpreter to synthesise all parameters simultaneously, mitigate the uncertainty of biased cut-off, and objectively classify lithofacies on the accuracy scale. In this paper, multiple classification algorithms including support vector machine (SVM), random forest (RF), decision tree (DT), K-nearest neighbours (KNN), logistic regression, Gaussian, Bernoulli, multinomial Naïve Bayes, and linear discriminant analysis were executed on the seismic attributes for lithofacies prediction. Initially, all data points of five seismic attributes of acoustic impedance, Lambda-Rho, Mu-Rho, density (ρ), and compressional wave to shear wave velocity (VpVs) within 25-metre radius and 25-metre interval offset top and base of reservoir were orbitally extracted on 4 wells to create the datasets. Cross-validation and grid search were also implemented on the best four algorithms to optimise the hyper-parameters for each algorithm and avoid overfitting during training. Finally, confusion matrix and accuracy scores were exploited to determine the ultimate model for discrete lithofacies prediction. The machine learning models were applied to predict lithofacies for a complex reservoir in an area of 163 km2. From the perspective of classification, the random forest method achieved the highest accuracy score of 0.907 compared to support vector machine (0.896), K-nearest neighbours (0.895), and decision tree (0.892). At well locations, the correlation factor was excellent with 0.88 for random forest results versus sand thickness. In terms of sand and shale distribution, the machine learning outputs demonstrated geologically reasonable results, even in undrilled regions and reservoir boundary areas.
{"title":"Supervised machine learning application of lithofacies classification for a hydrodynamically complex gas - condensate reservoir in Nam Con Son basin","authors":"N. Nguyen, Ngoc The Hung Tran, Ky Son Hoang, Vũ Tùng Trần","doi":"10.47800/pvj.2022.06-03","DOIUrl":"https://doi.org/10.47800/pvj.2022.06-03","url":null,"abstract":"Conventional integration of rock physics and seismic inversion can quantitatively evaluate and contrast reservoir properties. However, the available output attributes are occasionally not a perfect indicator for specific information such as lithology or fluid saturation due to technology constraints. Each attribute commonly exhibits a combination of geological characteristics that could lead to subjective interpretations and provides only qualitative results. Meanwhile, machine learning (ML) is emerging as an independent interpreter to synthesise all parameters simultaneously, mitigate the uncertainty of biased cut-off, and objectively classify lithofacies on the accuracy scale. \u0000In this paper, multiple classification algorithms including support vector machine (SVM), random forest (RF), decision tree (DT), K-nearest neighbours (KNN), logistic regression, Gaussian, Bernoulli, multinomial Naïve Bayes, and linear discriminant analysis were executed on the seismic attributes for lithofacies prediction. Initially, all data points of five seismic attributes of acoustic impedance, Lambda-Rho, Mu-Rho, density (ρ), and compressional wave to shear wave velocity (VpVs) within 25-metre radius and 25-metre interval offset top and base of reservoir were orbitally extracted on 4 wells to create the datasets. Cross-validation and grid search were also implemented on the best four algorithms to optimise the hyper-parameters for each algorithm and avoid overfitting during training. Finally, confusion matrix and accuracy scores were exploited to determine the ultimate model for discrete lithofacies prediction. The machine learning models were applied to predict lithofacies for a complex reservoir in an area of 163 km2. \u0000From the perspective of classification, the random forest method achieved the highest accuracy score of 0.907 compared to support vector machine (0.896), K-nearest neighbours (0.895), and decision tree (0.892). At well locations, the correlation factor was excellent with 0.88 for random forest results versus sand thickness. In terms of sand and shale distribution, the machine learning outputs demonstrated geologically reasonable results, even in undrilled regions and reservoir boundary areas.","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121320026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hữu Hải Ngô, Trọng Tài Nguyễn, Thế Vinh Nguyễn, Văn Từ Trương
The selection of drilling fluid systems for wells to be drilled plays a vital role in designing and operations, especially for wells that penetrate through challenging geological conditions or through reservoirs containing sour gas. For reservoirs with high CO2 content, during the drilling operations, CO2 gas from the reservoir will influx into the well which causes chemical reactions between CO2 and chemicals used in the mud systems, leading to changes of their properties such as density, viscosity and rheology. These changes lead to incorrect drilling parameters, which can cause problems during operations. The article proposes an option to select a suitable drilling fluid system for wells with high CO2 content in the Song Hong basin in order to reduce risks and improve the efficiency of drilling operations.
{"title":"Selection of drilling fluid system for wells with high CO2 content in Song Hong basin","authors":"Hữu Hải Ngô, Trọng Tài Nguyễn, Thế Vinh Nguyễn, Văn Từ Trương","doi":"10.47800/pvj.2022.05-02","DOIUrl":"https://doi.org/10.47800/pvj.2022.05-02","url":null,"abstract":"The selection of drilling fluid systems for wells to be drilled plays a vital role in designing and operations, especially for wells that penetrate through challenging geological conditions or through reservoirs containing sour gas. For reservoirs with high CO2 content, during the drilling operations, CO2 gas from the reservoir will influx into the well which causes chemical reactions between CO2 and chemicals used in the mud systems, leading to changes of their properties such as density, viscosity and rheology. These changes lead to incorrect drilling parameters, which can cause problems during operations. \u0000The article proposes an option to select a suitable drilling fluid system for wells with high CO2 content in the Song Hong basin in order to reduce risks and improve the efficiency of drilling operations. ","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130917991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hồng Nguyên Lê, Thị Tuyết Mai Đặng, Thị Bích Phương Đặng, Thị Ánh Trinh Lưu
Flue gas from gas turbines at Nhon Trach 1 and Nhon Trach 2 gas power plant are in the temperature range of 100 - 113oC after heat has been recovered at the heat recovery steam generator. These heat flows are not recovered by conventional methods since they are not effective. Meanwhile, the organic Rankine cycle (ORC) uses organic fluids with low boiling point, that is why it can recover heat from low-temperature flue gas streams. Results of the ORC investigation reveal that with R245fa as a fluid, the Nhon Trach 1’s capacity will increase by 2.0 MW, and the Nhon Trach 2’s capacity will see an increase of 3.6 MW with R113 as a fluid.
{"title":"Recovering heat of flue gas from heat recovery steam generator system at Nhon Trach 1 and Nhon Trach 2 gas power plants by organic Rankine cycle to produce power","authors":"Hồng Nguyên Lê, Thị Tuyết Mai Đặng, Thị Bích Phương Đặng, Thị Ánh Trinh Lưu","doi":"10.47800/pvj.2022.05-05","DOIUrl":"https://doi.org/10.47800/pvj.2022.05-05","url":null,"abstract":"Flue gas from gas turbines at Nhon Trach 1 and Nhon Trach 2 gas power plant are in the temperature range of 100 - 113oC after heat has been recovered at the heat recovery steam generator. These heat flows are not recovered by conventional methods since they are not effective. Meanwhile, the organic Rankine cycle (ORC) uses organic fluids with low boiling point, that is why it can recover heat from low-temperature flue gas streams. Results of the ORC investigation reveal that with R245fa as a fluid, the Nhon Trach 1’s capacity will increase by 2.0 MW, and the Nhon Trach 2’s capacity will see an increase of 3.6 MW with R113 as a fluid. ","PeriodicalId":294988,"journal":{"name":"Petrovietnam Journal","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126496088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：