Masahiro Nagao, A. Datta-Gupta, Tsubasa Onishi, S. Sankaran
Routine well-wise injection/production data contain significant information that can be used for closed-loop reservoir management and rapid field decision-making. Traditional physics-based numerical reservoir simulation can be computationally prohibitive for short-term decision cycles, and it requires a detailed geologic model. Reduced physics models provide an efficient simulator-free workflow but often have a limited range of applicability. Pure machine learning models lack physical interpretability and can have limited predictive power. To address these challenges, we propose hybrid models, combining machine learning and a physics-based approach, for rapid production forecasting and reservoir connectivity characterization using routine injection/production and pressure data. Our framework takes routine measurements, such as injection rate and pressure data, as inputs and multiphase production rates as outputs. We combine reduced physics models into a neural network architecture by utilizing two different approaches. In the first approach, the reduced physics model is used for preprocessing to obtain approximate solutions that feed it into a neural network as input. This physics-based input feature can reduce the model complexity and provide significant improvement in prediction performance. In the second approach, a physics-informed neural network (PINN) is applied. The residual terms are augmented in the neural network loss function as physics-based regularization that relies on the governing partial differential equations (PDE). Reduced physics models are used for the governing PDE to enable efficient neural network training. The regularization allows the model to avoid overfitting and provides better predictive performance. Our proposed hybrid models are first validated using a 2D benchmark reservoir simulation case and then applied to a field-scale reservoir case to show the robustness and efficiency of the method. The hybrid models are shown to provide prediction performance that is superior to pure machine learning models and reduced physics models in terms of multiphase production rates. Specifically, in the second method with PINN, the trained hybrid neural network model satisfies the reduced physics system, making it physically interpretable, and provides interwell connectivity in terms of well flux allocation. The flux allocation estimated from the hybrid model was compared with streamline-based flux allocation, and reasonable agreement was obtained. By combining the reduced physics model with the efficacy of deep learning, model calibration can be done very efficiently without constructing a geologic model. The proposed hybrid models with physics-based regularization and physics-based preprocessing provide novel approaches to augment data-driven models with underlying physics to build interpretable models for understanding reservoir connectivity between wells and for robust future production forecasting.
{"title":"Physics Informed Machine Learning for Reservoir Connectivity Identification and Robust Production Forecasting","authors":"Masahiro Nagao, A. Datta-Gupta, Tsubasa Onishi, S. Sankaran","doi":"10.2118/219773-pa","DOIUrl":"https://doi.org/10.2118/219773-pa","url":null,"abstract":"\u0000 Routine well-wise injection/production data contain significant information that can be used for closed-loop reservoir management and rapid field decision-making. Traditional physics-based numerical reservoir simulation can be computationally prohibitive for short-term decision cycles, and it requires a detailed geologic model. Reduced physics models provide an efficient simulator-free workflow but often have a limited range of applicability. Pure machine learning models lack physical interpretability and can have limited predictive power. To address these challenges, we propose hybrid models, combining machine learning and a physics-based approach, for rapid production forecasting and reservoir connectivity characterization using routine injection/production and pressure data.\u0000 Our framework takes routine measurements, such as injection rate and pressure data, as inputs and multiphase production rates as outputs. We combine reduced physics models into a neural network architecture by utilizing two different approaches. In the first approach, the reduced physics model is used for preprocessing to obtain approximate solutions that feed it into a neural network as input. This physics-based input feature can reduce the model complexity and provide significant improvement in prediction performance. In the second approach, a physics-informed neural network (PINN) is applied. The residual terms are augmented in the neural network loss function as physics-based regularization that relies on the governing partial differential equations (PDE). Reduced physics models are used for the governing PDE to enable efficient neural network training. The regularization allows the model to avoid overfitting and provides better predictive performance.\u0000 Our proposed hybrid models are first validated using a 2D benchmark reservoir simulation case and then applied to a field-scale reservoir case to show the robustness and efficiency of the method. The hybrid models are shown to provide prediction performance that is superior to pure machine learning models and reduced physics models in terms of multiphase production rates. Specifically, in the second method with PINN, the trained hybrid neural network model satisfies the reduced physics system, making it physically interpretable, and provides interwell connectivity in terms of well flux allocation. The flux allocation estimated from the hybrid model was compared with streamline-based flux allocation, and reasonable agreement was obtained. By combining the reduced physics model with the efficacy of deep learning, model calibration can be done very efficiently without constructing a geologic model.\u0000 The proposed hybrid models with physics-based regularization and physics-based preprocessing provide novel approaches to augment data-driven models with underlying physics to build interpretable models for understanding reservoir connectivity between wells and for robust future production forecasting.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"180 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141405574","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}
Boyun Guo, Md Nahin Mahmood, Philip B. Wortman, Vu V. Nguyen
Post-fracturing fluid soaking is believed to significantly affect productivity of hydraulic-fractured horizontal wells in the oil and gas industry. It is highly desirable to know the optimal soaking time period to maximize well productivity. An analytical model was developed in this research to describe the dynamic spontaneous imbibition process in shale cracks owing to fluid soaking. The model was integrated with pressure falloff data to develop a mathematical approach for forecasting the optimal soaking time, which is defined as the time it takes for the fluid to reach the midpoint between two adjacent hydraulic fractures. The optimal soaking time depends on crack and fluid properties, including crack width, fluid contact angle, and interfacial tension (IFT). A case study from the data of Tuscaloosa Marine Shale (TMS) suggests that the optimal soaking time is 2–4 weeks if the fracturing spacing is 4–6 m. This work provides petroleum engineers with a quick-and-easy method for predicting the optimal soaking time to maximize their well productivity.
{"title":"Prediction of the Optimal Post-Fracturing Soaking Time in Multifractured Shale Gas/Oil Formations on the Basis of Modeling of Fluid Imbibition","authors":"Boyun Guo, Md Nahin Mahmood, Philip B. Wortman, Vu V. Nguyen","doi":"10.2118/221459-pa","DOIUrl":"https://doi.org/10.2118/221459-pa","url":null,"abstract":"\u0000 Post-fracturing fluid soaking is believed to significantly affect productivity of hydraulic-fractured horizontal wells in the oil and gas industry. It is highly desirable to know the optimal soaking time period to maximize well productivity. An analytical model was developed in this research to describe the dynamic spontaneous imbibition process in shale cracks owing to fluid soaking. The model was integrated with pressure falloff data to develop a mathematical approach for forecasting the optimal soaking time, which is defined as the time it takes for the fluid to reach the midpoint between two adjacent hydraulic fractures. The optimal soaking time depends on crack and fluid properties, including crack width, fluid contact angle, and interfacial tension (IFT). A case study from the data of Tuscaloosa Marine Shale (TMS) suggests that the optimal soaking time is 2–4 weeks if the fracturing spacing is 4–6 m. This work provides petroleum engineers with a quick-and-easy method for predicting the optimal soaking time to maximize their well productivity.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"65 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141391107","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}
Depleted shale formations have the potential for hydrogen geostorage. The storage mechanisms, however, are complex and influenced by several factors including mineralogy, pore size distribution, residual hydrocarbons in place, and the choice of cushion gas. This study aims to investigate hydrogen distribution within this multiscale pore system, with a focus on understanding how hydrogen accumulates in the organic nanoporous network. Such insights are critical for the long-term storage and recovery assessments. Using molecular simulations, representative organic matter comprising nanoporous kerogen and nanopores of different sizes was constructed. Hydrogen intake of the organic system in the presence of residual amount of natural gas was quantified, considering multiple hydrogen injection scenarios. Despite stronger chemical affinity toward natural gas, hydrogen accumulated in all pore sizes, even the smallest, potentially beneficial for long-term storage but hindering rapid recovery. Moreover, the study was extended to investigate the role of cushion gas in the accumulation of hydrogen in organic structures. It was found that introducing cushion gases, such as methane and carbon dioxide, reduces hydrogen intake in the nanopores, with carbon dioxide being the most effective due to its stronger attraction to kerogen. Nitrogen, on the other hand, had relatively lower impact. The results were consistent with the observed trends in the analysis of the nonbonding energy of all systems. The results reported in this study provide critical insights into the factors influencing hydrogen accumulation in the organic constituents of shale formations for an optimized design of hydrogen geostorage in depleted shale gas reservoirs.
{"title":"Hydrogen Geostorage in Organic-Rich Shales: Critical Insights Into the Role of Cushion Gas in Hydrogen Accumulation within Organic Nanoporous Systems","authors":"Saad Alafnan","doi":"10.2118/221468-pa","DOIUrl":"https://doi.org/10.2118/221468-pa","url":null,"abstract":"\u0000 Depleted shale formations have the potential for hydrogen geostorage. The storage mechanisms, however, are complex and influenced by several factors including mineralogy, pore size distribution, residual hydrocarbons in place, and the choice of cushion gas. This study aims to investigate hydrogen distribution within this multiscale pore system, with a focus on understanding how hydrogen accumulates in the organic nanoporous network. Such insights are critical for the long-term storage and recovery assessments. Using molecular simulations, representative organic matter comprising nanoporous kerogen and nanopores of different sizes was constructed. Hydrogen intake of the organic system in the presence of residual amount of natural gas was quantified, considering multiple hydrogen injection scenarios. Despite stronger chemical affinity toward natural gas, hydrogen accumulated in all pore sizes, even the smallest, potentially beneficial for long-term storage but hindering rapid recovery. Moreover, the study was extended to investigate the role of cushion gas in the accumulation of hydrogen in organic structures. It was found that introducing cushion gases, such as methane and carbon dioxide, reduces hydrogen intake in the nanopores, with carbon dioxide being the most effective due to its stronger attraction to kerogen. Nitrogen, on the other hand, had relatively lower impact. The results were consistent with the observed trends in the analysis of the nonbonding energy of all systems. The results reported in this study provide critical insights into the factors influencing hydrogen accumulation in the organic constituents of shale formations for an optimized design of hydrogen geostorage in depleted shale gas reservoirs.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"83 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141391327","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}
Anoo Sebastian, Muhammad Mushtaq, E. Al-Shalabi, W. Alameri, K. Mohanty, S. K. Masalmeh, A. AlSumaiti
Polymer retention is considered a major challenge in polymer flooding applications, especially in carbonates. This is due to the prevailing conditions of low permeability (<100 md), high temperature (>85°C), and high salinity (>100,000 ppm) generally found in these formations, which limit the effectiveness of commonly used polymers such as hydrolyzed polyacrylamide (HPAM) and xanthan gum. To address these challenges, a polymer based on acrylamide tertiary butyl sulfonate (ATBS) has been used due to its tolerance to high-temperature and -salinity conditions. However, the high cost of manufacturing these polymers, combined with their anionic properties that promote adsorption onto positively charged carbonate rocks, necessitates the exploration of methods to reduce polymer retention. In this study, we aim to determine the sufficient concentration of hardness ions (Ca2+ and Mg2+) required to significantly reduce the adsorption of this polymer. The study is unique in its focus on mitigating polymer retention in carbonate formations using softened brine, as no prior research has investigated this aspect. Four different brines were investigated with a salinity of 8,000 ppm total dissolved salts (TDS) and varying ionic composition designed mainly by eliminating the hardness-causing ions, Ca2+ and Mg2+. A geochemical study was performed using the PHREEQC software to analyze the interaction between these injected brines and the rock. Furthermore, comprehensive rheological and static adsorption studies were performed at a temperature of 25°C using the potential ATBS-based polymer to evaluate the polymer performance and adsorption in these brines. Later, dynamic adsorption studies were conducted in both single-phase and two-phase conditions to further quantify polymer adsorption. The geochemical study showed an anhydrite saturation index (SI) of less than 0.5 for all the brines used when interacting with the rock, indicating a very low tendency for calcium sulfate precipitation. Furthermore, the rheological studies showed that polymer viscosity significantly increased with reduced hardness, where a polymer solution viscosity of 7.5 cp was obtained in zero hardness brine, nearly 1.5 times higher than the polymer viscosity of the base makeup brine of 8,000 ppm. Moreover, it was observed that, by carefully tuning the concentrations of the divalent cations, the polymer concentration consumption for the required target viscosity was reduced by 40–50%. For the single-phase static adsorption experiments, the polymer solution in softened brines resulted in lower adsorption in the range of 37–62 µg/g-rock as opposed to 102 µg/g-rock for the base makeup brine. On the other hand, the single-phase dynamic adsorption results showed an even lowered polymer adsorption of 33 µg/g-rock for the softened brine compared with 45 µg/g-rock for the base makeup brine. Additionally, the single-phase dynamic adsorption studies showed a remarkable improvement in polymer injectivit
{"title":"Investigating the Effect of Water Softening on Polymer Adsorption onto Carbonates through Single-Phase and Two-Phase Experiments","authors":"Anoo Sebastian, Muhammad Mushtaq, E. Al-Shalabi, W. Alameri, K. Mohanty, S. K. Masalmeh, A. AlSumaiti","doi":"10.2118/211470-pa","DOIUrl":"https://doi.org/10.2118/211470-pa","url":null,"abstract":"\u0000 Polymer retention is considered a major challenge in polymer flooding applications, especially in carbonates. This is due to the prevailing conditions of low permeability (<100 md), high temperature (>85°C), and high salinity (>100,000 ppm) generally found in these formations, which limit the effectiveness of commonly used polymers such as hydrolyzed polyacrylamide (HPAM) and xanthan gum. To address these challenges, a polymer based on acrylamide tertiary butyl sulfonate (ATBS) has been used due to its tolerance to high-temperature and -salinity conditions. However, the high cost of manufacturing these polymers, combined with their anionic properties that promote adsorption onto positively charged carbonate rocks, necessitates the exploration of methods to reduce polymer retention. In this study, we aim to determine the sufficient concentration of hardness ions (Ca2+ and Mg2+) required to significantly reduce the adsorption of this polymer. The study is unique in its focus on mitigating polymer retention in carbonate formations using softened brine, as no prior research has investigated this aspect. Four different brines were investigated with a salinity of 8,000 ppm total dissolved salts (TDS) and varying ionic composition designed mainly by eliminating the hardness-causing ions, Ca2+ and Mg2+. A geochemical study was performed using the PHREEQC software to analyze the interaction between these injected brines and the rock. Furthermore, comprehensive rheological and static adsorption studies were performed at a temperature of 25°C using the potential ATBS-based polymer to evaluate the polymer performance and adsorption in these brines. Later, dynamic adsorption studies were conducted in both single-phase and two-phase conditions to further quantify polymer adsorption.\u0000 The geochemical study showed an anhydrite saturation index (SI) of less than 0.5 for all the brines used when interacting with the rock, indicating a very low tendency for calcium sulfate precipitation. Furthermore, the rheological studies showed that polymer viscosity significantly increased with reduced hardness, where a polymer solution viscosity of 7.5 cp was obtained in zero hardness brine, nearly 1.5 times higher than the polymer viscosity of the base makeup brine of 8,000 ppm. Moreover, it was observed that, by carefully tuning the concentrations of the divalent cations, the polymer concentration consumption for the required target viscosity was reduced by 40–50%. For the single-phase static adsorption experiments, the polymer solution in softened brines resulted in lower adsorption in the range of 37–62 µg/g-rock as opposed to 102 µg/g-rock for the base makeup brine. On the other hand, the single-phase dynamic adsorption results showed an even lowered polymer adsorption of 33 µg/g-rock for the softened brine compared with 45 µg/g-rock for the base makeup brine. Additionally, the single-phase dynamic adsorption studies showed a remarkable improvement in polymer injectivit","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"17 11","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141411141","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}
When applying the high-pressure cyclic water injection technique in injection and production wells belonging to cracks-caverns reservoirs for huff and puff oil production, it is crucial to effectively judge the reservoir type, accurately calculate the reservoir parameters, and reasonably set the high-pressure water injection parameters, which can effectively solve the low recovery efficiency and rapid oil production decline of the injection and production wells due to the differences in cracks and caverns and its complexity in spatial development. However, the imperfection of the existing technical system, resulting in the inability to ensure the rationality and effectiveness of high-pressure cyclic water injection parameter settings, followed by the difficulty in evaluating oil increment of the injection and production wells after multiple rounds of water injection and oil production, greatly limit the deep application of this technique. To solve this problem, we take the Tahe Oilfield (a typical cracks-caverns reservoir in China) as an example. First, we analyze the morphological characteristics of numerous high-pressure water injection indicating curves in Tahe Oilfield, and extract four typical indicating curves using differentiated classification. Second, based on the volume balance equations, we establish two mathematical models—the karst cavern mathematical model of water injection indicating curve and the cracks-caverns mathematical model of water injection indicating curve. Finally, by solving the two mathematic models and the correlation analysis of characteristic parameters belonging to the four extracted typical indicating curves, we can fulfill the reservoir types identification and quantitative calculation of the key reservoir parameters in the injection and production wells. Application of this technique in Well TH1021XX indicates that its far-wellbore crude oil reserves are 69.80×104 m3, its activation pressure of the interconnected fractures ranges from 6.25 MPa to 8.25 MPa, and the error between the actual accumulated oil production and its predictive value is less than 4% after four rounds of high-pressure water injection and oil production. Meanwhile, the error assessment results of numerous wells are all within 10%, which validates the effectiveness and practicality of the research findings in this article.
{"title":"Study on the Mechanism of High-Pressure Cyclic Water Injection for Far-Wellbore Oil Extraction in Cracks-Caverns Reservoirs","authors":"Beibei Jiang, Guoqiang Zhang, Dong Wang, Jiabo Liu, Haitao Li, Hongwen Luo, Yong Chen, Dong Liu","doi":"10.2118/221464-pa","DOIUrl":"https://doi.org/10.2118/221464-pa","url":null,"abstract":"\u0000 When applying the high-pressure cyclic water injection technique in injection and production wells belonging to cracks-caverns reservoirs for huff and puff oil production, it is crucial to effectively judge the reservoir type, accurately calculate the reservoir parameters, and reasonably set the high-pressure water injection parameters, which can effectively solve the low recovery efficiency and rapid oil production decline of the injection and production wells due to the differences in cracks and caverns and its complexity in spatial development. However, the imperfection of the existing technical system, resulting in the inability to ensure the rationality and effectiveness of high-pressure cyclic water injection parameter settings, followed by the difficulty in evaluating oil increment of the injection and production wells after multiple rounds of water injection and oil production, greatly limit the deep application of this technique.\u0000 To solve this problem, we take the Tahe Oilfield (a typical cracks-caverns reservoir in China) as an example. First, we analyze the morphological characteristics of numerous high-pressure water injection indicating curves in Tahe Oilfield, and extract four typical indicating curves using differentiated classification. Second, based on the volume balance equations, we establish two mathematical models—the karst cavern mathematical model of water injection indicating curve and the cracks-caverns mathematical model of water injection indicating curve. Finally, by solving the two mathematic models and the correlation analysis of characteristic parameters belonging to the four extracted typical indicating curves, we can fulfill the reservoir types identification and quantitative calculation of the key reservoir parameters in the injection and production wells. Application of this technique in Well TH1021XX indicates that its far-wellbore crude oil reserves are 69.80×104 m3, its activation pressure of the interconnected fractures ranges from 6.25 MPa to 8.25 MPa, and the error between the actual accumulated oil production and its predictive value is less than 4% after four rounds of high-pressure water injection and oil production. Meanwhile, the error assessment results of numerous wells are all within 10%, which validates the effectiveness and practicality of the research findings in this article.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"31 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141396649","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}
T. L. B. Dias, M. A. Marins, C. L. Pagliari, R. M. E. Barbosa, M. D. De Campos, E. A. B. Silva, S. L. Netto
Fault detection and diagnosis are fundamental problems in the process of abnormal event detection in oil wells. This paper describes an open-source modular system that enables the efficient design of fault detectors and classifiers based on machine learning techniques. Events considered in this work are part of the publicly available 3W database developed by Petrobras, the Brazilian oil holding. Seven fault classes are considered, with distinct dynamics and patterns, as well as several instances of normal operation. We also show the effectiveness of the use of wavelet-based features, which provide multiscale time-frequency analysis, targeting a more realistic event modeling. A few challenges imposed by the 3W data set are addressed by combining both wavelet and statistical features, resulting in more accurate and more robust classifiers, with a 98.6% balanced accuracy in the multiclass problem, a significant improvement over the 94.2% previously reported in the literature.
{"title":"Development of Oilwell Fault Classifiers Using a Wavelet-Based Multivariable Approach in a Modular Architecture","authors":"T. L. B. Dias, M. A. Marins, C. L. Pagliari, R. M. E. Barbosa, M. D. De Campos, E. A. B. Silva, S. L. Netto","doi":"10.2118/221463-pa","DOIUrl":"https://doi.org/10.2118/221463-pa","url":null,"abstract":"\u0000 Fault detection and diagnosis are fundamental problems in the process of abnormal event detection in oil wells. This paper describes an open-source modular system that enables the efficient design of fault detectors and classifiers based on machine learning techniques. Events considered in this work are part of the publicly available 3W database developed by Petrobras, the Brazilian oil holding. Seven fault classes are considered, with distinct dynamics and patterns, as well as several instances of normal operation. We also show the effectiveness of the use of wavelet-based features, which provide multiscale time-frequency analysis, targeting a more realistic event modeling. A few challenges imposed by the 3W data set are addressed by combining both wavelet and statistical features, resulting in more accurate and more robust classifiers, with a 98.6% balanced accuracy in the multiclass problem, a significant improvement over the 94.2% previously reported in the literature.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"177 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141413633","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}
Liangliang Ding, Wenkang Chen, Chuanjun Han, Yongzhi Xue, Qisong Lei
The perforation-acidizing-testing combined technology has become the key technology for increasing the efficiency and speed of ultradeep well completion testing. However, the shock load and the wellbore pressure surge affect the stability and local strength of the lower packer string system during the perforation detonation. The energy generated by the perforation detonation is the fundamental source of the shock load and the wellbore pressure surge. The effect laws and distribution characteristics of the explosion energy of the perforating shaped charge is urgently needed. Therefore, a fluid-structure coupling method based on a structural-arbitrary Lagrangian-Euler algorithm (S-ALE) is used to construct a numerical model to forecast the explosion energy. The feasibility of the numerical model is verified by comparison with the field experimental results. The detailed studies on the output value and distribution characteristics of the explosion energy are carried out. The main control factors and influencing laws of the explosion energy are clarified. Then, an equation for the explosion energy prediction is fitted to lay the foundation for studying the wellbore pressure surge and the lower packer string system failure caused by the perforation detonation. The obtained results indicate that the explosion energy is mainly divided into three parts: the jet kinetic energy, the shell case energy, and the pressure surge energy. The pressure surge energy can reach 59.254 to 66.08%, the jet kinetic energy can reach 9.895 to 17.159%, and the shell case energy can reach 21.426 to 24.325%. The major sensitive parameters that affect the pressure surge energy are ranked as follows: the explosive mass, the explosive type, the shell thickness, the standoff distance, the cone angle of the liner, and the shot density. This work provides a reliable prediction method for the accurate description of the explosion energy conversion, which is critical for improving the success rate of the perforation-acidizing-testing combined technology.
{"title":"Study on Explosion Energy Conversion of a Perforating Shaped Charge during Perforation Detonation","authors":"Liangliang Ding, Wenkang Chen, Chuanjun Han, Yongzhi Xue, Qisong Lei","doi":"10.2118/219473-pa","DOIUrl":"https://doi.org/10.2118/219473-pa","url":null,"abstract":"\u0000 The perforation-acidizing-testing combined technology has become the key technology for increasing the efficiency and speed of ultradeep well completion testing. However, the shock load and the wellbore pressure surge affect the stability and local strength of the lower packer string system during the perforation detonation. The energy generated by the perforation detonation is the fundamental source of the shock load and the wellbore pressure surge. The effect laws and distribution characteristics of the explosion energy of the perforating shaped charge is urgently needed. Therefore, a fluid-structure coupling method based on a structural-arbitrary Lagrangian-Euler algorithm (S-ALE) is used to construct a numerical model to forecast the explosion energy. The feasibility of the numerical model is verified by comparison with the field experimental results. The detailed studies on the output value and distribution characteristics of the explosion energy are carried out. The main control factors and influencing laws of the explosion energy are clarified. Then, an equation for the explosion energy prediction is fitted to lay the foundation for studying the wellbore pressure surge and the lower packer string system failure caused by the perforation detonation. The obtained results indicate that the explosion energy is mainly divided into three parts: the jet kinetic energy, the shell case energy, and the pressure surge energy. The pressure surge energy can reach 59.254 to 66.08%, the jet kinetic energy can reach 9.895 to 17.159%, and the shell case energy can reach 21.426 to 24.325%. The major sensitive parameters that affect the pressure surge energy are ranked as follows: the explosive mass, the explosive type, the shell thickness, the standoff distance, the cone angle of the liner, and the shot density. This work provides a reliable prediction method for the accurate description of the explosion energy conversion, which is critical for improving the success rate of the perforation-acidizing-testing combined technology.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"24 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140082923","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}
We present artificial neural network (ANN) models for predicting the flowing bottomhole pressure (FBHP) of unconventional oil wells under gas lift operations. Well parameters, fluid properties, production/injection data, and bottomhole gauge pressures from 16 shale oil wells in Permian Basin, Texas, USA, are analyzed to determine key parameters affecting FBHP during the gas lift operation. For the reservoir fluid properties, several pressure-volume-temperature (PVT) models, such as Benedict-Webb-Rubin (BWR); Lee, Gonzalez, and Eakin; and Standing, among others, are examined against experimentally tuned fluid properties (i.e., viscosity, formation volume factor, and solution gas-oil ratio) to identify representative fluid (PVT) models for oil and gas properties. Pipe flow models (i.e., Hagedorn and Brown; Gray, Begs and Brill; and Petalas and Aziz) are also examined by comparing calculated FBHP against the bottomhole gauge pressures to identify a representative pipe flow model. Training and test data sets are then generated using the representative PVT and pipe flow models to develop a physics-based ANN model. The physics-based ANN model inputs are hydrocarbon fluid properties, liquid flow rate (qL), gas-liquid ratio (GLR), water-oil ratio (WOR), well true vertical depth (TVD), wellhead pressure (Pwh), wellhead temperature (Twh), and temperature gradient (dT/dh). A data-based ANN model is also developed based on only TVD, Pwh, qL, GLR, and WOR. Both physics- and data-based ANN models are trained through hyperparameter optimization using genetic algorithm and K-fold validation and then tested against the gauge FBHP. The results reveal that both models perform well with the FBHP prediction from field data with a normalized mean absolute error (NMAE) of around 10%. However, a comparison between results from the physics- and data-based ANN models shows that the accuracy of the physics-based model is higher at the later phase of the gas lift operation when the steady-state pipe flow is well established. On the contrary, the data-based model performs better for the early phase of gas lift operation when transient flow behavior is dominant. Developed ANN models and workflows can be applied to optimize gas lift operations under different fluid and well conditions.
我们提出了人工神经网络 (ANN) 模型，用于预测气举作业下非常规油井的流动井底压力 (FBHP)。我们分析了美国德克萨斯州二叠纪盆地 16 口页岩油井的油井参数、流体性质、生产/注入数据以及井底表压，以确定气举作业期间影响井底压力的关键参数。在储层流体属性方面，针对实验调整的流体属性（即粘度、地层体积因子和溶液气油比），研究了几种压力-体积-温度（PVT）模型，如 Benedict-Webb-Rubin (BWR)、Lee、Gonzalez 和 Eakin 以及 Standing 等，以确定油气属性的代表性流体（PVT）模型。还通过将计算的 FBHP 与井底表压进行比较来确定具有代表性的管流模型（即 Hagedorn 和 Brown；Gray、Begs 和 Brill；以及 Petalas 和 Aziz）。然后使用具有代表性的 PVT 和管流模型生成训练和测试数据集，以开发基于物理的 ANN 模型。基于物理的 ANN 模型输入包括碳氢化合物流体特性、液体流速 (qL)、气液比 (GLR)、水油比 (WOR)、油井实际垂直深度 (TVD)、井口压力 (Pwh)、井口温度 (Twh) 和温度梯度 (dT/dh)。此外，还开发了一个基于数据的 ANN 模型，该模型仅基于 TVD、Pwh、qL、GLR 和 WOR。使用遗传算法和 K-fold 验证，通过超参数优化对物理和数据 ANN 模型进行训练，然后根据仪器 FBHP 进行测试。结果表明，这两种模型都能很好地预测来自现场数据的 FBHP，归一化平均绝对误差 (NMAE) 约为 10%。然而，对基于物理的 ANN 模型和基于数据的 ANN 模型的结果进行比较后发现，基于物理的模型在气举运行的后期阶段，即稳态管道流量建立良好的阶段精度更高。相反，当瞬态流动行为占主导地位时，基于数据的模型在气举运行的早期阶段表现更好。开发的 ANN 模型和工作流程可用于优化不同流体和油井条件下的气举作业。
{"title":"Flowing Bottomhole Pressure during Gas Lift in Unconventional Oil Wells","authors":"Miao Jin, Hamid Emami‐Meybodi, Mohammad Ahmadi","doi":"10.2118/214832-pa","DOIUrl":"https://doi.org/10.2118/214832-pa","url":null,"abstract":"\u0000 We present artificial neural network (ANN) models for predicting the flowing bottomhole pressure (FBHP) of unconventional oil wells under gas lift operations. Well parameters, fluid properties, production/injection data, and bottomhole gauge pressures from 16 shale oil wells in Permian Basin, Texas, USA, are analyzed to determine key parameters affecting FBHP during the gas lift operation. For the reservoir fluid properties, several pressure-volume-temperature (PVT) models, such as Benedict-Webb-Rubin (BWR); Lee, Gonzalez, and Eakin; and Standing, among others, are examined against experimentally tuned fluid properties (i.e., viscosity, formation volume factor, and solution gas-oil ratio) to identify representative fluid (PVT) models for oil and gas properties. Pipe flow models (i.e., Hagedorn and Brown; Gray, Begs and Brill; and Petalas and Aziz) are also examined by comparing calculated FBHP against the bottomhole gauge pressures to identify a representative pipe flow model. Training and test data sets are then generated using the representative PVT and pipe flow models to develop a physics-based ANN model. The physics-based ANN model inputs are hydrocarbon fluid properties, liquid flow rate (qL), gas-liquid ratio (GLR), water-oil ratio (WOR), well true vertical depth (TVD), wellhead pressure (Pwh), wellhead temperature (Twh), and temperature gradient (dT/dh). A data-based ANN model is also developed based on only TVD, Pwh, qL, GLR, and WOR. Both physics- and data-based ANN models are trained through hyperparameter optimization using genetic algorithm and K-fold validation and then tested against the gauge FBHP. The results reveal that both models perform well with the FBHP prediction from field data with a normalized mean absolute error (NMAE) of around 10%. However, a comparison between results from the physics- and data-based ANN models shows that the accuracy of the physics-based model is higher at the later phase of the gas lift operation when the steady-state pipe flow is well established. On the contrary, the data-based model performs better for the early phase of gas lift operation when transient flow behavior is dominant. Developed ANN models and workflows can be applied to optimize gas lift operations under different fluid and well conditions.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"120 3-4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139880034","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}
Shuai Zhao, Wanfen Pu, Qingyuan Chen, C. Yuan, M. Varfolomeev
The in-situ combustion (ISC) technique has emerged as a significant approach for shale oil production. However, currently, there is a lack of experimental evidence supporting the stable propagation of combustion front within fractured shale. This study aimed to investigate the combustion characteristics within fractured shale by using a self-designed combustion tube (CT) and an experimental scheme. Subsequently, an analysis of shale structure and oil properties was conducted. The results demonstrated that while the combustion front could propagate through shale with a single fracture width of approximately 43 μm, the combustion intensity gradually diminished, leading to an inability to sustain stable propagation in the later part of the oil-detritus mixtures. The combustion intensity within the shale was enhanced by preheating the shale at 250°C, resulting in an improved oil recovery from 67.8% to 77.9%. The findings indicated that the complex fractured shale allowed for the stable propagation of the combustion front without a significant decrease in combustion intensity. Moreover, the T2 spectrum analysis of shale revealed a gradual expansion of the pore-fracture structure and improved shale connectivity after combustion. The T1-T2 response illustrated the transformation of solid and heavy components into lighter components. Furthermore, the content of saturates and H in the oil increased after combustion, whereas there was a significant decrease in resins, O, and S. Overall, this study provided technical evidence supporting the feasibility of employing the ISC technique for the development of shale oil reservoirs with additional fractures.
原地燃烧（ISC）技术已成为页岩油生产的重要方法。然而，目前还缺乏实验证据支持燃烧前沿在断裂页岩中的稳定传播。本研究旨在利用自行设计的燃烧管（CT）和实验方案，研究断裂页岩内的燃烧特性。随后，对页岩结构和石油特性进行了分析。结果表明，虽然燃烧前沿可以在单个断裂宽度约为 43 μm 的页岩中传播，但燃烧强度逐渐减弱，导致在油-杂质混合物的后期无法持续稳定传播。通过在 250°C 下预热页岩，提高了页岩内的燃烧强度，从而将采油率从 67.8% 提高到 77.9%。研究结果表明，复杂断裂页岩可使燃烧前沿稳定传播，而燃烧强度不会显著降低。此外，页岩的 T2 频谱分析表明，燃烧后孔隙-断裂结构逐渐扩大，页岩的连通性得到改善。T1-T2 反应表明固体和重组分转变为轻组分。此外，燃烧后石油中饱和物和 H 的含量增加，而树脂、O 和 S 的含量则显著减少。总之，这项研究提供了技术证据，支持采用 ISC 技术开发具有额外裂缝的页岩油藏的可行性。
{"title":"Propagation of Combustion Front within Fractured Shale and Its Influence on Shale Structure and Crude Oil Properties: An Experimental Study","authors":"Shuai Zhao, Wanfen Pu, Qingyuan Chen, C. Yuan, M. Varfolomeev","doi":"10.2118/219456-pa","DOIUrl":"https://doi.org/10.2118/219456-pa","url":null,"abstract":"\u0000 The in-situ combustion (ISC) technique has emerged as a significant approach for shale oil production. However, currently, there is a lack of experimental evidence supporting the stable propagation of combustion front within fractured shale. This study aimed to investigate the combustion characteristics within fractured shale by using a self-designed combustion tube (CT) and an experimental scheme. Subsequently, an analysis of shale structure and oil properties was conducted. The results demonstrated that while the combustion front could propagate through shale with a single fracture width of approximately 43 μm, the combustion intensity gradually diminished, leading to an inability to sustain stable propagation in the later part of the oil-detritus mixtures. The combustion intensity within the shale was enhanced by preheating the shale at 250°C, resulting in an improved oil recovery from 67.8% to 77.9%. The findings indicated that the complex fractured shale allowed for the stable propagation of the combustion front without a significant decrease in combustion intensity. Moreover, the T2 spectrum analysis of shale revealed a gradual expansion of the pore-fracture structure and improved shale connectivity after combustion. The T1-T2 response illustrated the transformation of solid and heavy components into lighter components. Furthermore, the content of saturates and H in the oil increased after combustion, whereas there was a significant decrease in resins, O, and S. Overall, this study provided technical evidence supporting the feasibility of employing the ISC technique for the development of shale oil reservoirs with additional fractures.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"41 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139880911","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}
Quantitative prediction of reservoir tectonic fracture development characteristics, opening pressures, and opening sequences is critical in the exploration and development of oil- and gas-bearing reservoirs and thus has received widespread attention. Using numerical simulations of the paleostress field during the formation of tectonic fractures and the rock fracture criterion, we predict the development and occurrence of fractures in the Middle Ordovician Yijianfang Formation in the Shunnan region of the Tarim Basin, China. The local paleostress fields reflected by the mechanical properties and occurrence of tectonic fractures obtained from core descriptions, acoustic emission (AE) experiments, paleomagnetic experiments, sound velocity measurements, and borehole breakouts were used to determine the regional paleostress and in-situ stress. We established a geomechanical model by combining the mechanical parameters of the rocks with the finite element method (FEM), optimizing the boundary conditions with a self-adaptive constraint algorithm, and conducting numerical simulations of the in-situ stresses. Fracture occurrence and numerical simulation results of the in-situ stress field were used to determine the opening pressure (Pk) and opening sequence of the fractures. The level of fracture development decreases away from the strike-slip fault in the study area. Fracture development is positively correlated with the Young’s modulus, paleostress difference, and paleostress difference coefficient of the rock. The direction of the maximum horizontal principal stress is from north-northeast (NNE) to northeast (NE). Initially, shear fractures and tensional fractures oriented NNE 30°–35° and NE 40°–45°, respectively, open during the water injection process. Pk is positively correlated with the horizontal stress difference coefficient and the angle between the fracture strike and the maximum horizontal principal stress. At the structural highs (burial depths shallower than 6450 m) and the structural lows (burial depths deeper than 6450 m), the burial depth correlates negatively and positively with Pk, respectively. This investigation of the development, occurrence, Pk, and opening sequence of tectonic fractures and their principal controlling factors will have a positive impact on the future exploration and production opportunities of similar fractured reservoirs.
储层构造裂缝发育特征、张开压力和张开序列的定量预测对含油气藏的勘探开发至关重要，因此受到广泛关注。利用构造裂缝形成过程中的古应力场数值模拟和岩石裂缝判据，我们预测了中国塔里木盆地顺南地区中奥陶统易家房组裂缝的发育和出现情况。利用岩芯描述、声发射（AE）实验、古地磁实验、声速测量和钻孔破口获得的力学性质和构造断裂发生所反映的局部古应力场，确定了区域古应力和原位应力。我们将岩石力学参数与有限元法（FEM）相结合，建立了地质力学模型，利用自适应约束算法优化了边界条件，并对原位应力进行了数值模拟。利用断裂的发生和原位应力场的数值模拟结果，确定了断裂的张开压力（Pk）和张开顺序。在研究区域，断裂发育程度在远离走向滑动断层的地方有所降低。断裂发育程度与岩石的杨氏模量、古应力差和古应力差系数呈正相关。最大水平主应力方向为北东北（NNE）至东北（NE）。最初，在注水过程中，方向分别为 NNE 30°-35°和 NE 40°-45°的剪切断裂和张拉断裂会打开。Pk 与水平应力差系数以及断裂走向与最大水平主应力之间的夹角呈正相关。在构造高位（埋深小于 6450 米）和构造低位（埋深大于 6450 米），埋深分别与 Pk 呈负相关和正相关。对构造裂缝的发育、出现、Pk 和张开顺序及其主要控制因素的研究，将对今后类似裂缝储层的勘探和生产机会产生积极影响。
{"title":"Quantitative Prediction of the Development and Opening Sequence of Fractures in an Ultradeep Carbonate Reservoir: A Case Study of the Middle Ordovician in the Shunnan Area, Tarim Basin, China","authors":"Yuntao Li, Wenlong Ding, Jun Han, Xuyun Chen, Cheng Huang, Jingtian Li, Shihao Ding","doi":"10.2118/219453-pa","DOIUrl":"https://doi.org/10.2118/219453-pa","url":null,"abstract":"\u0000 Quantitative prediction of reservoir tectonic fracture development characteristics, opening pressures, and opening sequences is critical in the exploration and development of oil- and gas-bearing reservoirs and thus has received widespread attention. Using numerical simulations of the paleostress field during the formation of tectonic fractures and the rock fracture criterion, we predict the development and occurrence of fractures in the Middle Ordovician Yijianfang Formation in the Shunnan region of the Tarim Basin, China. The local paleostress fields reflected by the mechanical properties and occurrence of tectonic fractures obtained from core descriptions, acoustic emission (AE) experiments, paleomagnetic experiments, sound velocity measurements, and borehole breakouts were used to determine the regional paleostress and in-situ stress. We established a geomechanical model by combining the mechanical parameters of the rocks with the finite element method (FEM), optimizing the boundary conditions with a self-adaptive constraint algorithm, and conducting numerical simulations of the in-situ stresses. Fracture occurrence and numerical simulation results of the in-situ stress field were used to determine the opening pressure (Pk) and opening sequence of the fractures. The level of fracture development decreases away from the strike-slip fault in the study area. Fracture development is positively correlated with the Young’s modulus, paleostress difference, and paleostress difference coefficient of the rock. The direction of the maximum horizontal principal stress is from north-northeast (NNE) to northeast (NE). Initially, shear fractures and tensional fractures oriented NNE 30°–35° and NE 40°–45°, respectively, open during the water injection process. Pk is positively correlated with the horizontal stress difference coefficient and the angle between the fracture strike and the maximum horizontal principal stress. At the structural highs (burial depths shallower than 6450 m) and the structural lows (burial depths deeper than 6450 m), the burial depth correlates negatively and positively with Pk, respectively. This investigation of the development, occurrence, Pk, and opening sequence of tectonic fractures and their principal controlling factors will have a positive impact on the future exploration and production opportunities of similar fractured reservoirs.","PeriodicalId":510854,"journal":{"name":"SPE Journal","volume":"19 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139818426","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：