Zahra Ashena, Hojjat Kabirzadeh, Jeong Woo Kim, Xin Wang, Mohammed Y. Ali
Summary By using a deep neural network (DNN), a novel technique is developed for a 2.5D joint inversion of gravity and magnetic anomalies to model subsurface salts and basement structures. The joint application of gravity and magnetic anomalies addresses the inherent nonuniqueness problem of geophysical inversions. Moreover, DNN is used to conduct the nonlinear inverse mapping of gravity and magnetic anomalies to depth-to-salt and depth-to-basement. To create the training data set, a three-layer forward model of the subsurface is designed indicating sediments, salts, and the basement. The length and height of the model are determined based on the dimensions of the target area to be investigated. Several random parameters are set to create different representations of the forward model by altering the depth and shape of the layers. Given the topography of the salts and basement layers as well as their predefined density and susceptibility values, the gravity and magnetic anomalies of the forward models are calculated. Using multiprocessing algorithms, thousands of training examples are simulated comprising gravity and magnetic anomalies as input features and depth-to-salt and depth-to-basement as labels. The application of the proposed technique is evaluated to interpret the salt–basement structures over hydrocarbon reservoirs in offshore United Arab Emirates (UAE). Correspondingly, a DNN model is trained using the simulated data set of the target region and is assessed by making predictions on the random actual and noise-added synthetic data. Finally, gravity-magnetic anomalies are fed into the DNN inverse model to estimate the salts and basement structures over three profiles. The results proved the capability of our technique in modeling the subsurface structures.
{"title":"Joint Inversion of Gravity and Magnetic Anomalies to Image Salt–Basement Structures Offshore Abu Dhabi, UAE, Using Deep Neural Networks","authors":"Zahra Ashena, Hojjat Kabirzadeh, Jeong Woo Kim, Xin Wang, Mohammed Y. Ali","doi":"10.2118/217982-pa","DOIUrl":"https://doi.org/10.2118/217982-pa","url":null,"abstract":"Summary By using a deep neural network (DNN), a novel technique is developed for a 2.5D joint inversion of gravity and magnetic anomalies to model subsurface salts and basement structures. The joint application of gravity and magnetic anomalies addresses the inherent nonuniqueness problem of geophysical inversions. Moreover, DNN is used to conduct the nonlinear inverse mapping of gravity and magnetic anomalies to depth-to-salt and depth-to-basement. To create the training data set, a three-layer forward model of the subsurface is designed indicating sediments, salts, and the basement. The length and height of the model are determined based on the dimensions of the target area to be investigated. Several random parameters are set to create different representations of the forward model by altering the depth and shape of the layers. Given the topography of the salts and basement layers as well as their predefined density and susceptibility values, the gravity and magnetic anomalies of the forward models are calculated. Using multiprocessing algorithms, thousands of training examples are simulated comprising gravity and magnetic anomalies as input features and depth-to-salt and depth-to-basement as labels. The application of the proposed technique is evaluated to interpret the salt–basement structures over hydrocarbon reservoirs in offshore United Arab Emirates (UAE). Correspondingly, a DNN model is trained using the simulated data set of the target region and is assessed by making predictions on the random actual and noise-added synthetic data. Finally, gravity-magnetic anomalies are fed into the DNN inverse model to estimate the salts and basement structures over three profiles. The results proved the capability of our technique in modeling the subsurface structures.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136169148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ziwei Liu, Yongfei Yang, Qi Zhang, Gloire Imani, Lei Zhang, Hai Sun, Junjie Zhong, Kai Zhang, Jun Yao
Summary The depletion of conventional reservoirs has led to increased interest in deep shale gas. Hydraulic fracturing addresses the challenge of developing low-permeability shale, involving hydro-mechanical coupling fracture propagation mechanics. Supercritical CO2 (SC-CO2) has become a promising alternative to fracturing fluids due to its ability to be buried underground after use. The high temperature, pressure, and stress of deep shale lead to the flow of fracturing fluid to plastic deformation of rock, resulting in microfractures. In this paper, we simulate the fracture propagation process of deep shale fractured by SC-CO2 based on the coupling of the Darcy-Brinkman-Biot method, which adopts the Navier-Stokes-like equation to solve the free flow region, and the Darcy equation with Biot’s theory to solve flow in the matrix. To clearly probe the mechanism of deep fracturing from a microscopic perspective, the plastic rock property is taken into consideration. We investigate the effects of injection velocity, rock plastic yield stress, formation pressure, and gas slippage effect on fluid saturation and fracture morphology, and find that increasing the injection rate of fracturing fluid can form better extended fractures and complex fracture networks, improving the fracturing effect. Furthermore, we find that it is more appropriate to adopt SC-CO2 as a fracturing fluid alternative in deep shale with higher plastic yield stress due to higher CO2 saturation in the matrix, indicating greater carbon sequestration potential. High confining pressure promotes the growth of shear fractures, which are capable of more complex fracture profiles. The gas slip effect has a significant impact on the stress field while ignoring the flow field. This study sheds light on which deep shale gas reservoirs are appropriate for the use of SC-CO2 as a fracturing fluid and offers recommendations for how to enhance the fracturing effect at the pore scale.
{"title":"Pore-Scale Simulation of Fracture Propagation by CO2 Flow Induced in Deep Shale Based on Hydro-Mechanical Coupled Model","authors":"Ziwei Liu, Yongfei Yang, Qi Zhang, Gloire Imani, Lei Zhang, Hai Sun, Junjie Zhong, Kai Zhang, Jun Yao","doi":"10.2118/217990-pa","DOIUrl":"https://doi.org/10.2118/217990-pa","url":null,"abstract":"Summary The depletion of conventional reservoirs has led to increased interest in deep shale gas. Hydraulic fracturing addresses the challenge of developing low-permeability shale, involving hydro-mechanical coupling fracture propagation mechanics. Supercritical CO2 (SC-CO2) has become a promising alternative to fracturing fluids due to its ability to be buried underground after use. The high temperature, pressure, and stress of deep shale lead to the flow of fracturing fluid to plastic deformation of rock, resulting in microfractures. In this paper, we simulate the fracture propagation process of deep shale fractured by SC-CO2 based on the coupling of the Darcy-Brinkman-Biot method, which adopts the Navier-Stokes-like equation to solve the free flow region, and the Darcy equation with Biot’s theory to solve flow in the matrix. To clearly probe the mechanism of deep fracturing from a microscopic perspective, the plastic rock property is taken into consideration. We investigate the effects of injection velocity, rock plastic yield stress, formation pressure, and gas slippage effect on fluid saturation and fracture morphology, and find that increasing the injection rate of fracturing fluid can form better extended fractures and complex fracture networks, improving the fracturing effect. Furthermore, we find that it is more appropriate to adopt SC-CO2 as a fracturing fluid alternative in deep shale with higher plastic yield stress due to higher CO2 saturation in the matrix, indicating greater carbon sequestration potential. High confining pressure promotes the growth of shear fractures, which are capable of more complex fracture profiles. The gas slip effect has a significant impact on the stress field while ignoring the flow field. This study sheds light on which deep shale gas reservoirs are appropriate for the use of SC-CO2 as a fracturing fluid and offers recommendations for how to enhance the fracturing effect at the pore scale.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135656219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Guohua Gao, Horacio Florez, Jeroen Vink, Carl Blom, Terence J. Wells, Jan Fredrik Edvard Saaf
Summary The Gauss-Newton (GN) trust region optimization methods perform robustly but may introduce significant overhead cost when using the conventional matrix factorization method to solve the associated GN trust region subproblem (GNTRS). Solving a GNTRS involves solving a nonlinear equation using an iterative Newton-Raphson (NR) method. In each NR iteration, a symmetric linear system can be solved by different matrix factorization methods, including Cholesky decomposition (CD), eigenvalue decomposition (EVD), and singular value decomposition (SVD). Because CD fails to factorize a singular symmetric matrix, we propose solving a GNTRS using the robust EVD method. In this paper, we analyze the performances of different methods to solve a GNTRS using different matrix factorization subroutines in LAPACK with different options and settings. The cost of solving a GNTRS mainly depends on the number of observed data (m) and the number of uncertainty parameters (n). When n≤m, we recommend directly solving the original GNTRS with n variables. When n>m, we propose an indirect method that transforms the original GNTRS with n variables to a new problem with m unknowns. The proposed indirect method can significantly reduce the computational cost by dimension reduction. However, dimension reduction may introduce numerical errors, which, in turn, may result in accuracy degradation and cause failure of convergence using the popular iterative NR method. To further improve the overall performance, we introduce a numerical error indicator to terminate the iterative NR process when numerical errors become dominant. Finally, we benchmarked the performances of different approaches on a set of testing problems with different settings. Our results confirm that the GNTRS solver using the EVD method together with the modified NR method performs the best, being both robust (no failure for all testing problems) and efficient (consuming comparable CPU time to other methods).
{"title":"Performance Benchmarking of Different Methods to Solve Gauss-Newton Trust Region Subproblems","authors":"Guohua Gao, Horacio Florez, Jeroen Vink, Carl Blom, Terence J. Wells, Jan Fredrik Edvard Saaf","doi":"10.2118/212180-pa","DOIUrl":"https://doi.org/10.2118/212180-pa","url":null,"abstract":"Summary The Gauss-Newton (GN) trust region optimization methods perform robustly but may introduce significant overhead cost when using the conventional matrix factorization method to solve the associated GN trust region subproblem (GNTRS). Solving a GNTRS involves solving a nonlinear equation using an iterative Newton-Raphson (NR) method. In each NR iteration, a symmetric linear system can be solved by different matrix factorization methods, including Cholesky decomposition (CD), eigenvalue decomposition (EVD), and singular value decomposition (SVD). Because CD fails to factorize a singular symmetric matrix, we propose solving a GNTRS using the robust EVD method. In this paper, we analyze the performances of different methods to solve a GNTRS using different matrix factorization subroutines in LAPACK with different options and settings. The cost of solving a GNTRS mainly depends on the number of observed data (m) and the number of uncertainty parameters (n). When n≤m, we recommend directly solving the original GNTRS with n variables. When n>m, we propose an indirect method that transforms the original GNTRS with n variables to a new problem with m unknowns. The proposed indirect method can significantly reduce the computational cost by dimension reduction. However, dimension reduction may introduce numerical errors, which, in turn, may result in accuracy degradation and cause failure of convergence using the popular iterative NR method. To further improve the overall performance, we introduce a numerical error indicator to terminate the iterative NR process when numerical errors become dominant. Finally, we benchmarked the performances of different approaches on a set of testing problems with different settings. Our results confirm that the GNTRS solver using the EVD method together with the modified NR method performs the best, being both robust (no failure for all testing problems) and efficient (consuming comparable CPU time to other methods).","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135964044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yuan Mingjian, Jun Liu, Zhigang Du, Zhengzhi Zhou, Kai Tang
Summary With the vigorous development of unconventional oil and gas reservoirs, logging technology of horizontal-well drillpipes has become increasingly important. The improper selection of displacement during the pumping of drilling rods into horizontal wells can lead to accidents such as cable entanglement or tool-string pump detachment. Based on a comprehensive consideration of multiple factors such as drillpipe joints, small gap flow, pressure loss along the pipe, temperature, pressure, tool-string passability, and waterhole diameter variation, a cable tension analysis model for a logging tool string of a horizontal-well drillpipe is established. This model can predict a reasonable pumping displacement by analyzing the tension in the entire wellbore section, thereby avoiding cable entanglement or even tool-string pump detachment accidents during downhole operation. To verify the effectiveness of the model, we analyzed the cable tension during logging operations in Wells M1 and M2 using on-site test data and compared them with the measured tension data. The determination coefficient (R2), mean absolute error (MAE), and root mean square error (RMSE) were used to evaluate the degree of fit and accuracy of the model. The experimental results showed that the model can accurately predict the pumping displacement range before logging operations, ensuring that the cable does not entangle and the tool string is not pumped out during the downhole operation, thus providing effective methods and theoretical guidance for on-site drilling rod-pumping logging operations.
{"title":"Research on a Calculation Model of Cable Tension and Pumping Displacement of a Logging Tool String of Horizontal-Well Drillpipes","authors":"Yuan Mingjian, Jun Liu, Zhigang Du, Zhengzhi Zhou, Kai Tang","doi":"10.2118/217984-pa","DOIUrl":"https://doi.org/10.2118/217984-pa","url":null,"abstract":"Summary With the vigorous development of unconventional oil and gas reservoirs, logging technology of horizontal-well drillpipes has become increasingly important. The improper selection of displacement during the pumping of drilling rods into horizontal wells can lead to accidents such as cable entanglement or tool-string pump detachment. Based on a comprehensive consideration of multiple factors such as drillpipe joints, small gap flow, pressure loss along the pipe, temperature, pressure, tool-string passability, and waterhole diameter variation, a cable tension analysis model for a logging tool string of a horizontal-well drillpipe is established. This model can predict a reasonable pumping displacement by analyzing the tension in the entire wellbore section, thereby avoiding cable entanglement or even tool-string pump detachment accidents during downhole operation. To verify the effectiveness of the model, we analyzed the cable tension during logging operations in Wells M1 and M2 using on-site test data and compared them with the measured tension data. The determination coefficient (R2), mean absolute error (MAE), and root mean square error (RMSE) were used to evaluate the degree of fit and accuracy of the model. The experimental results showed that the model can accurately predict the pumping displacement range before logging operations, ensuring that the cable does not entangle and the tool string is not pumped out during the downhole operation, thus providing effective methods and theoretical guidance for on-site drilling rod-pumping logging operations.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136093441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Summary The relative permeability expresses the mobility reduction factor when a fluid flows through a porous medium in the presence of another fluid and appears in Darcy’s law for multiphase flow. In this work, we replace Darcy’s law with more general momentum equations accounting for fluid-rock interaction (flow resistance), fluid-fluid interaction (drag), and Brinkman terms responding to gradients in fluid interstitial velocities. By coupling the momentum equations with phase transport equations, we study two important flow processes—forced imbibition (coreflooding) and countercurrent spontaneous imbibition. In the former, a constant water injection rate is applied and capillary forces are neglected, while in the latter, capillary forces drive the process and the total flux is zero. Our aim is to understand what relative permeabilities result from these systems and flow configurations. From previous work, when using momentum equations without Brinkman terms, unique saturation-dependent relative permeabilities are obtained for the two flow modes that depend on the flow mode. Now, with Brinkman terms included, the relative permeabilities depend on local spatial derivatives of interstitial velocity and pressure. Local relative permeabilities are calculated for both phases utilizing the ratio of phase Darcy velocity and phase pressure gradient. In addition, we use the Johnson-Bossler-Naumann (JBN) method for forced imbibition (with data simulated under the assumption of negligible capillary end effects) to calculate interpreted relative permeabilities from pressure drop and average saturation. Both flow setups are parameterized with literature data, and sensitivity analysis is performed. During coreflooding, Brinkman terms give a flatter saturation profile and higher front saturation. The saturation profile shape changes with time. Local water relative permeabilities are reduced, while they are slightly raised for oil. The saturation range where relative permeabilities can be evaluated locally is raised and made narrower with increased Brinkman terms. JBN relative permeabilities deviate from the local values: The trends in curves and saturation range are the same but more pronounced as they incorporate average measurements, including the strong impact at the inlet. Brinkman effects vanish after sufficient distance traveled, resulting in the unique saturation functions as a limit. Unsteady state (USS) relative permeabilities (based on transient data from single-phase injection) differ from steady-state (SS) relative permeabilities (based on SS data from coinjection of two fluids) because the Brinkman terms are zero at SS. During spontaneous imbibition, the higher effect from the Brinkman terms caused oil relative permeabilities to decrease at low water saturations and slightly increase at high saturations, while water relative permeability was only slightly reduced. The net effect was a delay in the imbibition profile. Local relative permeabilities
{"title":"Effective Relative Permeabilities Based on Momentum Equations with Brinkman Terms and Viscous Coupling","authors":"Yangyang Qiao, Pål Østebø Andersen","doi":"10.2118/214388-pa","DOIUrl":"https://doi.org/10.2118/214388-pa","url":null,"abstract":"Summary The relative permeability expresses the mobility reduction factor when a fluid flows through a porous medium in the presence of another fluid and appears in Darcy’s law for multiphase flow. In this work, we replace Darcy’s law with more general momentum equations accounting for fluid-rock interaction (flow resistance), fluid-fluid interaction (drag), and Brinkman terms responding to gradients in fluid interstitial velocities. By coupling the momentum equations with phase transport equations, we study two important flow processes—forced imbibition (coreflooding) and countercurrent spontaneous imbibition. In the former, a constant water injection rate is applied and capillary forces are neglected, while in the latter, capillary forces drive the process and the total flux is zero. Our aim is to understand what relative permeabilities result from these systems and flow configurations. From previous work, when using momentum equations without Brinkman terms, unique saturation-dependent relative permeabilities are obtained for the two flow modes that depend on the flow mode. Now, with Brinkman terms included, the relative permeabilities depend on local spatial derivatives of interstitial velocity and pressure. Local relative permeabilities are calculated for both phases utilizing the ratio of phase Darcy velocity and phase pressure gradient. In addition, we use the Johnson-Bossler-Naumann (JBN) method for forced imbibition (with data simulated under the assumption of negligible capillary end effects) to calculate interpreted relative permeabilities from pressure drop and average saturation. Both flow setups are parameterized with literature data, and sensitivity analysis is performed. During coreflooding, Brinkman terms give a flatter saturation profile and higher front saturation. The saturation profile shape changes with time. Local water relative permeabilities are reduced, while they are slightly raised for oil. The saturation range where relative permeabilities can be evaluated locally is raised and made narrower with increased Brinkman terms. JBN relative permeabilities deviate from the local values: The trends in curves and saturation range are the same but more pronounced as they incorporate average measurements, including the strong impact at the inlet. Brinkman effects vanish after sufficient distance traveled, resulting in the unique saturation functions as a limit. Unsteady state (USS) relative permeabilities (based on transient data from single-phase injection) differ from steady-state (SS) relative permeabilities (based on SS data from coinjection of two fluids) because the Brinkman terms are zero at SS. During spontaneous imbibition, the higher effect from the Brinkman terms caused oil relative permeabilities to decrease at low water saturations and slightly increase at high saturations, while water relative permeability was only slightly reduced. The net effect was a delay in the imbibition profile. Local relative permeabilities","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135567701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Youssuf Elnoamany, Andreas Michael, Ipsita Gupta, Tej Bhinde, Steve Todman, Paulo J. Waltrich, Yuanhang Chen
Summary Failed well-capping attempts following offshore-well blowouts undergoing worst case discharge (WCD) can lead to fluid-driven tensile failures (de facto hydraulic fracture initiations). Subsequent propagation of these fractures may lead to broaching of overburden-rock-formation layers and even the seafloor, providing pathways for reservoir hydrocarbons to escape into the seawater. After capping stack shut-in, the pressure buildup in the fluid column inside the wellbore exposes vulnerable locations to tensile (Mode I) failure. If this shut-in wellbore pressure exceeds the formation breakdown pressure (FBP) in any of the exposed-rock-formation layers, a fracture will initiate and will continue to propagate as long as sufficient energy is provided by the reservoir. Scenarios where the propagating fracture(s) broached the seafloor in the past led to severe environmental impacts, disturbing the local ecology. The quintessential example is Union Oil’s 1969 “A-21 Well” blowout in California’s Santa Barbara Channel, where subsequent well-capping attempts led to multiple broaching instances on the seafloor near the well with thousands of hydrocarbon gallons gushing into the seawater (observable from the sea surface as “oil boilups”). In this paper, numerical modeling is performed on a hypothetical case study using deepwater Gulf of Mexico parameters in order to evaluate the likelihood of a similar scenario by modeling a planar-fracture propagation longitudinally-to-the-wellbore, upon well capping. A workflow is developed that integrates post-blowout WCD flowrate and volume estimations, fracture initiation and propagation modeling following the capping stack shut-in, pertaining to a “cap-and-contain” strategy (including predictions in regard to seafloor broaching), along with relief-well intersection followed by kill-weight-mud injection. The casing-shoe depth is the presumed location of fracture initiation, assuming perfect integrity of all casedhole sections above it. Several sensitivity analyses are performed to investigate the impact of the casing-shoe depth, along with the stiffness of the overburden-rock-formation layer, and the post-blowout-discharge duration on the resultant fracture growth. Finally, the mud density and pump flowrate necessary to compensate the oil column to successfully kill the blown well are quantitatively assessed.
{"title":"Blowout-Capping-Fracturing-Relief Well: A Full Cycle Workflow","authors":"Youssuf Elnoamany, Andreas Michael, Ipsita Gupta, Tej Bhinde, Steve Todman, Paulo J. Waltrich, Yuanhang Chen","doi":"10.2118/217462-pa","DOIUrl":"https://doi.org/10.2118/217462-pa","url":null,"abstract":"Summary Failed well-capping attempts following offshore-well blowouts undergoing worst case discharge (WCD) can lead to fluid-driven tensile failures (de facto hydraulic fracture initiations). Subsequent propagation of these fractures may lead to broaching of overburden-rock-formation layers and even the seafloor, providing pathways for reservoir hydrocarbons to escape into the seawater. After capping stack shut-in, the pressure buildup in the fluid column inside the wellbore exposes vulnerable locations to tensile (Mode I) failure. If this shut-in wellbore pressure exceeds the formation breakdown pressure (FBP) in any of the exposed-rock-formation layers, a fracture will initiate and will continue to propagate as long as sufficient energy is provided by the reservoir. Scenarios where the propagating fracture(s) broached the seafloor in the past led to severe environmental impacts, disturbing the local ecology. The quintessential example is Union Oil’s 1969 “A-21 Well” blowout in California’s Santa Barbara Channel, where subsequent well-capping attempts led to multiple broaching instances on the seafloor near the well with thousands of hydrocarbon gallons gushing into the seawater (observable from the sea surface as “oil boilups”). In this paper, numerical modeling is performed on a hypothetical case study using deepwater Gulf of Mexico parameters in order to evaluate the likelihood of a similar scenario by modeling a planar-fracture propagation longitudinally-to-the-wellbore, upon well capping. A workflow is developed that integrates post-blowout WCD flowrate and volume estimations, fracture initiation and propagation modeling following the capping stack shut-in, pertaining to a “cap-and-contain” strategy (including predictions in regard to seafloor broaching), along with relief-well intersection followed by kill-weight-mud injection. The casing-shoe depth is the presumed location of fracture initiation, assuming perfect integrity of all casedhole sections above it. Several sensitivity analyses are performed to investigate the impact of the casing-shoe depth, along with the stiffness of the overburden-rock-formation layer, and the post-blowout-discharge duration on the resultant fracture growth. Finally, the mud density and pump flowrate necessary to compensate the oil column to successfully kill the blown well are quantitatively assessed.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135606891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Summary Annular pressure buildup due to fluid expansion can be mitigated by using a compressible fluid, typically nitrogen, as a cushion at the top of an annuli. The advantage of using a nitrogen cushion is that we do not have to manipulate annuli pressures as often during variations in production. A disadvantage is that it is more difficult to detect small leaks to or from an annulus. For gas lift-assisted production wells, Annulus A is used for the transportation of gas down to the gas lift valves (GLVs), effectively making up a large gas cushion compared with the full length for the annulus. In light of this, monitoring annular pressures and ensuring continuous control of fluid volumes are essential for effective well barrier management. We present relevant theory and show that we can track annuli liquid levels using distributed temperature sensing (DTS) and/or distributed acoustic sensing (DAS) data to detect leaks, estimate leak rates, and infer leak paths. We find that the main cause for observing liquid levels in these data is because the equilibrium temperature at the fiber is dependent on the fluid fill of the various annuli in addition to the temperature inside the tubing and outside of the well. Six data examples with variations in liquid level(s) are presented to demonstrate this. Furthermore, simple models for estimating changes in liquid levels are proposed and compared with liquid levels from distributed fiber-optic (FO) data. Being able to detect leaks to or from annuli makes it possible for the operator to apply mitigating action in a timely manner, prevent unwanted well integrity situations, and ensure production regularity.
{"title":"Annuli Liquid-Level Surveillance Using Distributed Fiber-Optic Sensing Data","authors":"Kjetil E. Haavik","doi":"10.2118/217989-pa","DOIUrl":"https://doi.org/10.2118/217989-pa","url":null,"abstract":"Summary Annular pressure buildup due to fluid expansion can be mitigated by using a compressible fluid, typically nitrogen, as a cushion at the top of an annuli. The advantage of using a nitrogen cushion is that we do not have to manipulate annuli pressures as often during variations in production. A disadvantage is that it is more difficult to detect small leaks to or from an annulus. For gas lift-assisted production wells, Annulus A is used for the transportation of gas down to the gas lift valves (GLVs), effectively making up a large gas cushion compared with the full length for the annulus. In light of this, monitoring annular pressures and ensuring continuous control of fluid volumes are essential for effective well barrier management. We present relevant theory and show that we can track annuli liquid levels using distributed temperature sensing (DTS) and/or distributed acoustic sensing (DAS) data to detect leaks, estimate leak rates, and infer leak paths. We find that the main cause for observing liquid levels in these data is because the equilibrium temperature at the fiber is dependent on the fluid fill of the various annuli in addition to the temperature inside the tubing and outside of the well. Six data examples with variations in liquid level(s) are presented to demonstrate this. Furthermore, simple models for estimating changes in liquid levels are proposed and compared with liquid levels from distributed fiber-optic (FO) data. Being able to detect leaks to or from annuli makes it possible for the operator to apply mitigating action in a timely manner, prevent unwanted well integrity situations, and ensure production regularity.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135663613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Marcelo J. Dall’Aqua, Emilio J. R. Coutinho, Eduardo Gildin, Zhenyu Guo, Hardik Zalavadia, Sathish Sankaran
Summary Integrated reservoir studies for performance prediction and decision-making processes are computationally expensive. In this paper, we develop a novel linearization approach to reduce the computational burden of intensive reservoir simulation execution. We achieve this by introducing two novel components: (1) augmention of the state-space to yield a bilinear system and (2) an autoencoder based on a deep neural network to linearize physics reservoir equations in a reduced manifold using a Koopman operator. Recognizing that reservoir simulators execute expensive Newton-Raphson iterations after each timestep to solve the nonlinearities of the physical model, we propose “lifting” the physics to a more amenable manifold where the model behaves close to a linear system, similar to the Koopman theory, thus avoiding the iteration step. We use autoencoder deep neural networks with specific loss functions and structure to transform the nonlinear equation and frame it as a bilinear system with constant matrices over time. In such a way, it forces the states (pressures and saturations) to evolve in time by simple matrix multiplications in the lifted manifold. We also adopt a “guided” training approach, which is performed in three steps: (1) We initially train the autoencoder, (2) then we use a “conventional” model order reduction (MOR) as an initializer for the final (3) full training, when we use reservoir knowledge to improve and to lead the results to physically meaningful output. Many simulation studies exhibit extremely nonlinear and multiscale behavior, which can be difficult to model and control. Koopman operators can be shown to represent any dynamical system through linear dynamics. We applied this new framework to a 2D two-phase (oil and water) reservoir subject to a waterflooding plan with three wells (one injector and two producers) with speedups around 100 times faster and accuracy in the order of 1% to 3% on the pressure and saturation predictions. It is worthwhile noting that this method is a nonintrusive data-driven method because it does not need access to the reservoir simulation internal structure; thus, it is easily applied to commercial reservoir simulators and is also extendable to other studies. In addition, an extra benefit of this framework is to enable the plethora of well-developed tools for MOR of linear systems. To the authors’ knowledge, this is the first work that uses the Koopman operator for linearizing the system with controls. As with any MOR method, this can be directly applied to a well-control optimization problem and well-placement studies with low computational cost in the prediction step and good accuracy.
{"title":"Guided Deep Learning Manifold Linearization of Porous Media Flow Equations","authors":"Marcelo J. Dall’Aqua, Emilio J. R. Coutinho, Eduardo Gildin, Zhenyu Guo, Hardik Zalavadia, Sathish Sankaran","doi":"10.2118/212204-pa","DOIUrl":"https://doi.org/10.2118/212204-pa","url":null,"abstract":"Summary Integrated reservoir studies for performance prediction and decision-making processes are computationally expensive. In this paper, we develop a novel linearization approach to reduce the computational burden of intensive reservoir simulation execution. We achieve this by introducing two novel components: (1) augmention of the state-space to yield a bilinear system and (2) an autoencoder based on a deep neural network to linearize physics reservoir equations in a reduced manifold using a Koopman operator. Recognizing that reservoir simulators execute expensive Newton-Raphson iterations after each timestep to solve the nonlinearities of the physical model, we propose “lifting” the physics to a more amenable manifold where the model behaves close to a linear system, similar to the Koopman theory, thus avoiding the iteration step. We use autoencoder deep neural networks with specific loss functions and structure to transform the nonlinear equation and frame it as a bilinear system with constant matrices over time. In such a way, it forces the states (pressures and saturations) to evolve in time by simple matrix multiplications in the lifted manifold. We also adopt a “guided” training approach, which is performed in three steps: (1) We initially train the autoencoder, (2) then we use a “conventional” model order reduction (MOR) as an initializer for the final (3) full training, when we use reservoir knowledge to improve and to lead the results to physically meaningful output. Many simulation studies exhibit extremely nonlinear and multiscale behavior, which can be difficult to model and control. Koopman operators can be shown to represent any dynamical system through linear dynamics. We applied this new framework to a 2D two-phase (oil and water) reservoir subject to a waterflooding plan with three wells (one injector and two producers) with speedups around 100 times faster and accuracy in the order of 1% to 3% on the pressure and saturation predictions. It is worthwhile noting that this method is a nonintrusive data-driven method because it does not need access to the reservoir simulation internal structure; thus, it is easily applied to commercial reservoir simulators and is also extendable to other studies. In addition, an extra benefit of this framework is to enable the plethora of well-developed tools for MOR of linear systems. To the authors’ knowledge, this is the first work that uses the Koopman operator for linearizing the system with controls. As with any MOR method, this can be directly applied to a well-control optimization problem and well-placement studies with low computational cost in the prediction step and good accuracy.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"119 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135962861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Summary Cracking in the cement sheath of oil and gas wells is a major concern because it compromises well integrity and can lead to uncontrolled leaks of hydrocarbons, affecting both well safety and the environment. Among several reasons that might induce cracking, in the present study, we focus on the radial expansion of the steel casing resulting from pressure changes as one specific cause of damage. A test setup was designed to mechanically expand the steel casing while cameras monitored the cement to detect and measure cracking using digital image correlation (DIC) techniques. Six full-size replicas of 9-5/8-in. oilwell cross-sections were tested, and cracks in the cement ranging from 10 µm to 500 µm in width were quantified. Although each specimen exhibited a unique cracking pattern without a clear trend in the measured crack widths, analysis of the crack areas revealed a distinct pattern. Across all specimens, the cracked area showed (i) rapid growth at casing radial expansions between 0 µm and 100 µm, reaching cracked area values around 15 mm²; (ii) a gradually slower increase at casing radial expansions between 100 µm and 250 µm, reaching cracked areas up to 25 mm²; and (iii) a relatively constant cracked area stabilizing at approximately 25 mm² beyond radial expansions of 250 µm.
{"title":"Quantification of Casing Expansion-Induced Cracking in Oilwell Cement Sheaths","authors":"Pablo Alberdi-Pagola, Gregor Fischer","doi":"10.2118/218003-pa","DOIUrl":"https://doi.org/10.2118/218003-pa","url":null,"abstract":"Summary Cracking in the cement sheath of oil and gas wells is a major concern because it compromises well integrity and can lead to uncontrolled leaks of hydrocarbons, affecting both well safety and the environment. Among several reasons that might induce cracking, in the present study, we focus on the radial expansion of the steel casing resulting from pressure changes as one specific cause of damage. A test setup was designed to mechanically expand the steel casing while cameras monitored the cement to detect and measure cracking using digital image correlation (DIC) techniques. Six full-size replicas of 9-5/8-in. oilwell cross-sections were tested, and cracks in the cement ranging from 10 µm to 500 µm in width were quantified. Although each specimen exhibited a unique cracking pattern without a clear trend in the measured crack widths, analysis of the crack areas revealed a distinct pattern. Across all specimens, the cracked area showed (i) rapid growth at casing radial expansions between 0 µm and 100 µm, reaching cracked area values around 15 mm²; (ii) a gradually slower increase at casing radial expansions between 100 µm and 250 µm, reaching cracked areas up to 25 mm²; and (iii) a relatively constant cracked area stabilizing at approximately 25 mm² beyond radial expansions of 250 µm.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136159656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Summary Due to a dipping and folded structure, hydrate-bearing sediments (HBS) have obvious fluctuation characteristics, and the internal temperature and pressure of HBS are unevenly distributed. Subsequently, gas and water production of natural gas hydrate (NGH) is affected. When using a numerical simulation method to predict effectively the productivity of HBS, it is necessary to establish a conceptual model that considers the formation fluctuation. However, few reported studies accurately describe the fluctuation characteristics of HBS in numerical simulation models. Therefore, the spatial evolution of gas production, water production, and seepage parameters of each model was compared by establishing the initial temperature and pressure model of each representative model pair, using the TOUGH + HYDRATE (T + H) code for a long-time simulation; the production process of gas and water and spatial evolution of seepage parameters of each model were compared; and then the spatial evolution of gas production, water production, and seepage parameters of the different dipping/folded HBS was obtained. The spatial evolution of water production and seepage parameters for different dipping/folded HBS is obtained. (a) The dipping and folded structure had an obvious influence on the spatial distribution of the initial temperature and pressure of HBS. (b) The limits of heat supply and seepage capacity of the fluctuating HBS gave lower gas production than in horizontal HBS. There should be more emphasis on heat supply conditions and the formation of secondary hydrates. (c) The additional pore water in fluctuating HBS is not conducive to the discharge of methane. Consequently, the development of improved water blocking measures is significant for the future large-scale production of NGH.
{"title":"Effects of Dipping and Folded Structure on Gas Production from Hydrate-Bearing Sediments","authors":"Yaobin Li, Tianfu Xu, Xin Xin, Yingqi Zang, Han Yu, Huixing Zhu, Yilong Yuan","doi":"10.2118/217991-pa","DOIUrl":"https://doi.org/10.2118/217991-pa","url":null,"abstract":"Summary Due to a dipping and folded structure, hydrate-bearing sediments (HBS) have obvious fluctuation characteristics, and the internal temperature and pressure of HBS are unevenly distributed. Subsequently, gas and water production of natural gas hydrate (NGH) is affected. When using a numerical simulation method to predict effectively the productivity of HBS, it is necessary to establish a conceptual model that considers the formation fluctuation. However, few reported studies accurately describe the fluctuation characteristics of HBS in numerical simulation models. Therefore, the spatial evolution of gas production, water production, and seepage parameters of each model was compared by establishing the initial temperature and pressure model of each representative model pair, using the TOUGH + HYDRATE (T + H) code for a long-time simulation; the production process of gas and water and spatial evolution of seepage parameters of each model were compared; and then the spatial evolution of gas production, water production, and seepage parameters of the different dipping/folded HBS was obtained. The spatial evolution of water production and seepage parameters for different dipping/folded HBS is obtained. (a) The dipping and folded structure had an obvious influence on the spatial distribution of the initial temperature and pressure of HBS. (b) The limits of heat supply and seepage capacity of the fluctuating HBS gave lower gas production than in horizontal HBS. There should be more emphasis on heat supply conditions and the formation of secondary hydrates. (c) The additional pore water in fluctuating HBS is not conducive to the discharge of methane. Consequently, the development of improved water blocking measures is significant for the future large-scale production of NGH.","PeriodicalId":22252,"journal":{"name":"SPE Journal","volume":"238 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136160814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","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：