D. Mishra, Suraj Kumar, Vineet Mishra, Mohan Lal, V. Avadhani
In Mumbai offshore, Miocene carbonates are deposited with intermediate clastic inputs under cyclic sea level changes and have undergone diagenesis from time to time. Miocene carbonate layers deposited southwest of Mumbai High are producing a good amount of hydrocarbon from 1 to 2 Ω·m resistivity pays. A total of 58 representative core plugs from four different wells were studied to identify the reason for low resistivity and to classify rock facies types and porosity systems using scanning electron microscopy (SEM), thin-section nuclear magnetic resonance (NMR), and petrophysical core data. It was observed from the core study that Miocene carbonates have complex porosity systems and mud-supported to grain-supported reservoir facies. Dominance of mud-supported matrix is the main reason for low resistivity in Miocene carbonate layers as observed from integrated advanced log and core studies. Conventional petrophysical evaluation using constant petrophysical parameters (a, m, n) or linear correlation of cementation factor with porosity can lead to erroneous results in this scenario. A petrofacies-dependent correlation among cementation factor and porosity is attempted in this study for realistic evaluation of low-resistivity carbonate reservoirs. Different cementation factors vs. porosity relations have been derived for various carbonate formations worldwide. Shell formula demonstrates that cementation factor increases with decreasing porosity while correlation derived by Borai and Rafiee brought out inverse relation among cementation factors with porosity in tight carbonates and is providing almost constant cementation factor beyond 0.2. But, in our study, a core porosity-cementation factor plot of reservoir facies is showing that below 0.1, m values are increasing with increase of porosity, which is contradictory to Shell formula. This trend of cementation factor at low porosities is due to the presence of secondary porosity. In the porosity range 0.1–0.25, cementation factor increases eventually with the increase of porosity, but beyond porosity values 0.25, increase in porosity causes decrease of cementation factor. This is due to increasing content of mud-supported matrix, which is overall increasing the total porosity but eventually decreasing cementation in a rock. A new nonlinear correlation has been established between m and porosity for Miocene carbonates of Mumbai offshore area, by incorporating all the factors affecting cementation factor (m). Finally, saturation estimated using variable m either using newly established core derived correlation or resistivity image data is giving representative and improved saturation against low-resistivity reservoir layers compared with constant m.
{"title":"Petrofacies-Dependent Cementation Factor Relationship for Low-Resistivity Miocene Carbonates: Mumbai Offshore Basin","authors":"D. Mishra, Suraj Kumar, Vineet Mishra, Mohan Lal, V. Avadhani","doi":"10.2118/212854-pa","DOIUrl":"https://doi.org/10.2118/212854-pa","url":null,"abstract":"\u0000 In Mumbai offshore, Miocene carbonates are deposited with intermediate clastic inputs under cyclic sea level changes and have undergone diagenesis from time to time. Miocene carbonate layers deposited southwest of Mumbai High are producing a good amount of hydrocarbon from 1 to 2 Ω·m resistivity pays. A total of 58 representative core plugs from four different wells were studied to identify the reason for low resistivity and to classify rock facies types and porosity systems using scanning electron microscopy (SEM), thin-section nuclear magnetic resonance (NMR), and petrophysical core data. It was observed from the core study that Miocene carbonates have complex porosity systems and mud-supported to grain-supported reservoir facies. Dominance of mud-supported matrix is the main reason for low resistivity in Miocene carbonate layers as observed from integrated advanced log and core studies. Conventional petrophysical evaluation using constant petrophysical parameters (a, m, n) or linear correlation of cementation factor with porosity can lead to erroneous results in this scenario. A petrofacies-dependent correlation among cementation factor and porosity is attempted in this study for realistic evaluation of low-resistivity carbonate reservoirs. Different cementation factors vs. porosity relations have been derived for various carbonate formations worldwide. Shell formula demonstrates that cementation factor increases with decreasing porosity while correlation derived by Borai and Rafiee brought out inverse relation among cementation factors with porosity in tight carbonates and is providing almost constant cementation factor beyond 0.2. But, in our study, a core porosity-cementation factor plot of reservoir facies is showing that below 0.1, m values are increasing with increase of porosity, which is contradictory to Shell formula. This trend of cementation factor at low porosities is due to the presence of secondary porosity. In the porosity range 0.1–0.25, cementation factor increases eventually with the increase of porosity, but beyond porosity values 0.25, increase in porosity causes decrease of cementation factor. This is due to increasing content of mud-supported matrix, which is overall increasing the total porosity but eventually decreasing cementation in a rock. A new nonlinear correlation has been established between m and porosity for Miocene carbonates of Mumbai offshore area, by incorporating all the factors affecting cementation factor (m). Finally, saturation estimated using variable m either using newly established core derived correlation or resistivity image data is giving representative and improved saturation against low-resistivity reservoir layers compared with constant m.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"43 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87390752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this study, we developed a data mining-based multivariate analysis (MVA) workflow to identify correlations in complex high-dimensional data sets of small size. The research was motivated by the integration analysis of geologic, geophysical, completion, and production data from a 4-square-mile study field located in the Northern Denver-Julesburg (DJ) Basin, Colorado, USA. The goal is to establish a workflow that can extract learnings from a small data set to guide the future development of surrounding acreages. In this research, we propose an MVA workflow, which is modified significantly based on the random forest algorithm and assessed using the R2 score from K-fold cross-validation (CV). The MVA workflow performs significantly better in small data sets compared to traditional feature selection methods. This is because the MVA workflow includes (1) the selection of top-performing feature combinations at each step, (2) iterations embedded, (3) avoidance of random correlation, and (4) the summarization of each feature’s occurrence at the end. When the MVA workflow was initially applied on a complex synthetic small data set that included numerical and categorical variables, linear and nonlinear relationships, relationships within independent variables, and high dimensionality, it correctly identified all correlating variables and outperformed traditional feature selection methods. Following that, a field data set consisting of the information from 23 wells was investigated using the MVA workflow aiming at identifying the key factors that affect the production performance in the study area. The MVA workflow reveals the weak correlation between production and legacy well effect. The results show that the key factors affecting production in this study area are total organic carbon (TOC) percentage, open fracture densities, clay content, and legacy well effect, which should receive significant attention when developing neighboring acreage of the DJ Basin. More importantly, this MVA method can be implemented in other basins. Considering the heterogeneity of unconventional resources, it is worthwhile to identify the key production drivers on a small scale. The outperformance of this MVA method on small data sets makes it possible to provide valuable insights for each specific acreage.
{"title":"Application of Data Mining to Small Data Sets: Identification of Key Production Drivers in Heterogeneous Unconventional Resources","authors":"Yanrui Ning, H. Schumann, G. Jin","doi":"10.2118/212301-pa","DOIUrl":"https://doi.org/10.2118/212301-pa","url":null,"abstract":"\u0000 In this study, we developed a data mining-based multivariate analysis (MVA) workflow to identify correlations in complex high-dimensional data sets of small size. The research was motivated by the integration analysis of geologic, geophysical, completion, and production data from a 4-square-mile study field located in the Northern Denver-Julesburg (DJ) Basin, Colorado, USA. The goal is to establish a workflow that can extract learnings from a small data set to guide the future development of surrounding acreages. In this research, we propose an MVA workflow, which is modified significantly based on the random forest algorithm and assessed using the R2 score from K-fold cross-validation (CV). The MVA workflow performs significantly better in small data sets compared to traditional feature selection methods. This is because the MVA workflow includes (1) the selection of top-performing feature combinations at each step, (2) iterations embedded, (3) avoidance of random correlation, and (4) the summarization of each feature’s occurrence at the end. When the MVA workflow was initially applied on a complex synthetic small data set that included numerical and categorical variables, linear and nonlinear relationships, relationships within independent variables, and high dimensionality, it correctly identified all correlating variables and outperformed traditional feature selection methods. Following that, a field data set consisting of the information from 23 wells was investigated using the MVA workflow aiming at identifying the key factors that affect the production performance in the study area. The MVA workflow reveals the weak correlation between production and legacy well effect. The results show that the key factors affecting production in this study area are total organic carbon (TOC) percentage, open fracture densities, clay content, and legacy well effect, which should receive significant attention when developing neighboring acreage of the DJ Basin. More importantly, this MVA method can be implemented in other basins. Considering the heterogeneity of unconventional resources, it is worthwhile to identify the key production drivers on a small scale. The outperformance of this MVA method on small data sets makes it possible to provide valuable insights for each specific acreage.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"134 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86321336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Commerciality assessment and reserves booking is an important task, routinely performed by every oil and gas company. The task is performed based on the guidelines presented in the Petroleum Resources Management System (PRMS). The PRMS, revised in 2018, is a system developed for consistent and reliable definition, classification, and estimation of hydrocarbon resources. The document was prepared by the Oil and Gas Reserves Committee under the auspices of the Society of Petroleum Engineers. The PRMS has been broadly adopted by the oil and gas industry as an international standard reference for the classification and categorization of oil and gas reserves and resources. The objective of this paper is to reiterate the well-known knowledge regarding the economic viability, economic limit, sunk cost, depreciation, incremental project vs. entity’s economics, economic contingency, and technical contingency in a simple, unambiguous, and concise manner. The paper emphasizes using only discounted cash flows in all economic evaluations whether it is for reserves booking or investment decision-making. It is unfortunate that this well-known knowledge is incorrectly described and used in the industry for commerciality assessment of oil and gas resources and investment decision-making. The paper also shows that some of the financial assumptions and definitions such as financing at the incremental project level, properly accounting for sunk cost, and treatment of income tax when an entity has positive taxable income, if applied correctly, can greatly and legally enhance the chances of commerciality and reserves booking.
{"title":"Impact of Financial Assumptions on Economic Viability of Incremental Projects","authors":"M. A. Mian","doi":"10.2118/212826-pa","DOIUrl":"https://doi.org/10.2118/212826-pa","url":null,"abstract":"\u0000 Commerciality assessment and reserves booking is an important task, routinely performed by every oil and gas company. The task is performed based on the guidelines presented in the Petroleum Resources Management System (PRMS). The PRMS, revised in 2018, is a system developed for consistent and reliable definition, classification, and estimation of hydrocarbon resources. The document was prepared by the Oil and Gas Reserves Committee under the auspices of the Society of Petroleum Engineers. The PRMS has been broadly adopted by the oil and gas industry as an international standard reference for the classification and categorization of oil and gas reserves and resources.\u0000 The objective of this paper is to reiterate the well-known knowledge regarding the economic viability, economic limit, sunk cost, depreciation, incremental project vs. entity’s economics, economic contingency, and technical contingency in a simple, unambiguous, and concise manner. The paper emphasizes using only discounted cash flows in all economic evaluations whether it is for reserves booking or investment decision-making. It is unfortunate that this well-known knowledge is incorrectly described and used in the industry for commerciality assessment of oil and gas resources and investment decision-making.\u0000 The paper also shows that some of the financial assumptions and definitions such as financing at the incremental project level, properly accounting for sunk cost, and treatment of income tax when an entity has positive taxable income, if applied correctly, can greatly and legally enhance the chances of commerciality and reserves booking.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"2 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87164487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Elsayed, A. El-Husseiny, S. R. Hussaini, Hani Al Mukainah, Mohamed Mahmoud
Nuclear magnetic resonance (NMR) is a reliable tool for petrophysical evaluation and the characterization of pore structures. Compared to conventional carbonate reservoirs, sandstone reservoirs contain higher amounts of ferromagnetic and paramagnetic ions (such as iron, nickel, or manganese) usually found in microporous clay aggregates. The interpretation of petrophysical data in sandstone formations can be complicated by variations in clay mineralogy and distribution patterns (laminated, structural, and dispersed). Nevertheless, the impact of clay distribution patterns on NMR signals is not well understood. This study aims to investigate the impact of clay mineralogy and distribution patterns on the T2 relaxation times and internal gradient (i.e., inhomogeneity in the magnetic field). Glass beads were mixed with three different clay minerals characterized by no iron content (kaolinite) to higher iron content (illite and nontronite). The bead-clay mixtures were prepared at a fixed clay content but with variable clay distribution patterns to examine the impact of clay distribution alone. NMR T2 measurements at several echo times were performed on the pure glass beads and the mixtures to evaluate how clay mineralogy and distribution patterns affect the T2 and the internal gradient of the host glass beads. At a given clay distribution pattern, a more significant decrease in T2 relaxation times and a larger increase in the internal gradient of glass beads were observed when adding clays with higher iron content. This is explained by the higher surface relaxivity, and magnetic susceptibility caused when introducing clay with higher iron content. Such an impact can complicate the characterization of NMR-derived pore sizes as similar pore size distribution (PSD) can have very different T2 distribution and the logarithmic mean of T2 relaxation time distribution (T2LM) values. Micro-computed tomography (µCT) images were acquired to compute the PSD to compare it with ones obtained from the NMR measurements. The PSD for the three clay minerals showed almost the same distribution using µCT; however, they showed totally different T2 relaxation times distributions. That is due to the significant impact of the internal gradient causing a distortion in the magnetic field. Thus, careful consideration must be taken before converting the NMR data into PSD. The introduction of iron-free kaolinite resulted in a negligible impact on the internal gradient of glass beads regardless of the clay distribution pattern. On the other hand, the addition of dispersed iron-rich clays (illite and nontronite) results in up to two orders of magnitude increase in internal gradients, compared to mixtures with laminated and structural clays (at the same clay mineralogy and content). Moreover, dispersed clay mixtures display larger changes in T2LM and porosity as a function of increasing echo time. The results from this study suggest that changes in T2LM and the logarithmic mean of t
{"title":"Experimental Study on the Impact of Clay Distribution and Mineralogy on NMR T2 and Internal Gradient: Insights From Well-Controlled Physical Samples","authors":"M. Elsayed, A. El-Husseiny, S. R. Hussaini, Hani Al Mukainah, Mohamed Mahmoud","doi":"10.2118/212300-pa","DOIUrl":"https://doi.org/10.2118/212300-pa","url":null,"abstract":"\u0000 Nuclear magnetic resonance (NMR) is a reliable tool for petrophysical evaluation and the characterization of pore structures. Compared to conventional carbonate reservoirs, sandstone reservoirs contain higher amounts of ferromagnetic and paramagnetic ions (such as iron, nickel, or manganese) usually found in microporous clay aggregates. The interpretation of petrophysical data in sandstone formations can be complicated by variations in clay mineralogy and distribution patterns (laminated, structural, and dispersed). Nevertheless, the impact of clay distribution patterns on NMR signals is not well understood. This study aims to investigate the impact of clay mineralogy and distribution patterns on the T2 relaxation times and internal gradient (i.e., inhomogeneity in the magnetic field). Glass beads were mixed with three different clay minerals characterized by no iron content (kaolinite) to higher iron content (illite and nontronite). The bead-clay mixtures were prepared at a fixed clay content but with variable clay distribution patterns to examine the impact of clay distribution alone. NMR T2 measurements at several echo times were performed on the pure glass beads and the mixtures to evaluate how clay mineralogy and distribution patterns affect the T2 and the internal gradient of the host glass beads. At a given clay distribution pattern, a more significant decrease in T2 relaxation times and a larger increase in the internal gradient of glass beads were observed when adding clays with higher iron content. This is explained by the higher surface relaxivity, and magnetic susceptibility caused when introducing clay with higher iron content. Such an impact can complicate the characterization of NMR-derived pore sizes as similar pore size distribution (PSD) can have very different T2 distribution and the logarithmic mean of T2 relaxation time distribution (T2LM) values. Micro-computed tomography (µCT) images were acquired to compute the PSD to compare it with ones obtained from the NMR measurements. The PSD for the three clay minerals showed almost the same distribution using µCT; however, they showed totally different T2 relaxation times distributions. That is due to the significant impact of the internal gradient causing a distortion in the magnetic field. Thus, careful consideration must be taken before converting the NMR data into PSD. The introduction of iron-free kaolinite resulted in a negligible impact on the internal gradient of glass beads regardless of the clay distribution pattern. On the other hand, the addition of dispersed iron-rich clays (illite and nontronite) results in up to two orders of magnitude increase in internal gradients, compared to mixtures with laminated and structural clays (at the same clay mineralogy and content). Moreover, dispersed clay mixtures display larger changes in T2LM and porosity as a function of increasing echo time. The results from this study suggest that changes in T2LM and the logarithmic mean of t","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"18 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86095095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Fassihi, A. Alamatsaz, R. Moore, S. A. (Raj) Mehta, M. Ursenbach, D. Mallory, P. Pereira Almao, S. C. Gupta, H. Chhina
To understand the role of connate water as a source of hydrogen in oxidation and upgrading of bituminous oil at high temperature, heavy water (D2O) and O-18 enriched water (H2O18) were used as connate water in two different in-situ combustion experiments using a conical tube. Aside from fundamental understanding of the role of such reactions in in-situ combustion, the results could also potentially help in optimizing in-situ hydrogen generation and upgrading of heavy and bituminous oil. The conical tube had previously been used for understanding the impact of air flux (AF) in sustaining the combustion front (Alamatsaz et al. 2011). Significant upgrading was observed in these tests with a produced API gravity of 35 ° compared to the original bitumen gravity of 9 °API. This paper (Part A) deals with the experimental results including a comparison between conical tube and a combustion tube (CT) results. A subsequent paper (Part B) will discuss the upgrading aspects and the mass spectrometry results.
为了解原生水在烟油高温氧化升级过程中作为氢源的作用，利用锥形管进行了两种不同的原位燃烧实验，分别以重水(D2O)和O-18富集水(H2O18)作为原生水。除了对这些反应在原位燃烧中的作用有了基本的了解外，这些结果还可能有助于优化原位制氢和重质油和沥青油的升级。锥形管以前被用于了解空气通量(AF)对维持燃烧前沿的影响(Alamatsaz et al. 2011)。在这些测试中，与原始沥青重力为9°的API相比，生产的API重力达到35°，效果显著。本文(A部分)讨论了实验结果，包括锥形管和燃烧管(CT)结果的比较。随后的论文(B部分)将讨论升级方面和质谱分析结果。
{"title":"A Novel Approach in Understanding the Role of Water in Oxidation and Upgrading Reactions during In-Situ Combustion Oil Recovery, Part A: Experimental Observations","authors":"M. Fassihi, A. Alamatsaz, R. Moore, S. A. (Raj) Mehta, M. Ursenbach, D. Mallory, P. Pereira Almao, S. C. Gupta, H. Chhina","doi":"10.2118/212277-pa","DOIUrl":"https://doi.org/10.2118/212277-pa","url":null,"abstract":"\u0000 To understand the role of connate water as a source of hydrogen in oxidation and upgrading of bituminous oil at high temperature, heavy water (D2O) and O-18 enriched water (H2O18) were used as connate water in two different in-situ combustion experiments using a conical tube. Aside from fundamental understanding of the role of such reactions in in-situ combustion, the results could also potentially help in optimizing in-situ hydrogen generation and upgrading of heavy and bituminous oil. The conical tube had previously been used for understanding the impact of air flux (AF) in sustaining the combustion front (Alamatsaz et al. 2011). Significant upgrading was observed in these tests with a produced API gravity of 35 ° compared to the original bitumen gravity of 9 °API. This paper (Part A) deals with the experimental results including a comparison between conical tube and a combustion tube (CT) results. A subsequent paper (Part B) will discuss the upgrading aspects and the mass spectrometry results.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"54 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76664396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Luciana Maria da Silva, Leandro M. Ferreira, G. Avansi, D. Schiozer, S. N. Alves-Souza
One of the challenges related to reservoir engineering studies is working with essential high-dimensional inputs, such as porosity and permeability, which govern fluid flow in porous media. Dimensionality reduction (DR) methods have enabled spatial variability in constructing a fast objective function estimator (FOFE). This study presents a methodology to select an adequate DR method to deal with high-dimensional spatial attributes with more than 105 dimensions. We investigated 18 methods of DR commonly applied in the literature. The proposed workflow accomplished (1) definition of the adequate number of dimensions; (2) evaluation of the time spent for each data set generated using the elapsed computational time; (3) training using the automated machine learning (AutoML) technique; (4) validation using the root mean square logarithmic error (RMSLE) and the confidence interval (CI) of 95%; (5) a score equation using elapsed computational time and RMSLE; and (6) consistency check to evaluate if the FOFE is reliable to mimic simulator output. We used FOFE to generate risk curves at the final forecast period (10,957 days) as an application. We obtained methods that reduced the high-dimensional spatial attributes with a computational time lower than 10 minutes, enabling us to consider them in the FOFE building. We could deal with high-dimensional spatial variability from those selected approaches. Moreover, we can use the DR method selected to deal with high complexity problems to build an FOFE and avoid overfitting when a massive number of data are used.
{"title":"Selection of a Dimensionality Reduction Method: An Application to Deal with High-Dimensional Geostatistical Realizations in Oil Reservoirs","authors":"Luciana Maria da Silva, Leandro M. Ferreira, G. Avansi, D. Schiozer, S. N. Alves-Souza","doi":"10.2118/212299-pa","DOIUrl":"https://doi.org/10.2118/212299-pa","url":null,"abstract":"\u0000 One of the challenges related to reservoir engineering studies is working with essential high-dimensional inputs, such as porosity and permeability, which govern fluid flow in porous media. Dimensionality reduction (DR) methods have enabled spatial variability in constructing a fast objective function estimator (FOFE). This study presents a methodology to select an adequate DR method to deal with high-dimensional spatial attributes with more than 105 dimensions. We investigated 18 methods of DR commonly applied in the literature. The proposed workflow accomplished (1) definition of the adequate number of dimensions; (2) evaluation of the time spent for each data set generated using the elapsed computational time; (3) training using the automated machine learning (AutoML) technique; (4) validation using the root mean square logarithmic error (RMSLE) and the confidence interval (CI) of 95%; (5) a score equation using elapsed computational time and RMSLE; and (6) consistency check to evaluate if the FOFE is reliable to mimic simulator output. We used FOFE to generate risk curves at the final forecast period (10,957 days) as an application. We obtained methods that reduced the high-dimensional spatial attributes with a computational time lower than 10 minutes, enabling us to consider them in the FOFE building. We could deal with high-dimensional spatial variability from those selected approaches. Moreover, we can use the DR method selected to deal with high complexity problems to build an FOFE and avoid overfitting when a massive number of data are used.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"15 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74435035","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Well placement optimization is one of the most crucial tasks in the petroleum industry. It often involves high risk in the presence of geological uncertainty due to a limited understanding of the subsurface reservoir. Well placement optimization is different from decision selection as countless alternatives are impossible to be enumerated in a decision model (such as the mean-variance model). In many practical applications, the decision criterion of well placement optimization is based on maximizing the risk-adjusted value (mean-variance optimization) to capture different risk attitudes. This approach regards variance as the measure of risk, and it is performed under the expected utility framework. However, investors only dislike the downside volatility below a certain benchmark. The downside-risk approach has been discussed in previous studies, in this paper, it will be introduced in the well placement optimization and discussed under the expected utility framework. It is demonstrated in a synthetic reservoir model with the consideration of spatial heterogeneity, and the comparison between the downside-risk optimization and mean-variance optimization is also presented in this example. The observation implies that well placement optimization is heavily influenced by individuals’ preference to risk. The downside-risk optimization outperforms the mean-variance optimization because it explicitly assesses risk and does not penalize high outcomes.
{"title":"Managing Risk in Well Placement Optimization within an Expected Utility Framework","authors":"Di Yang, C. Deutsch","doi":"10.2118/212305-pa","DOIUrl":"https://doi.org/10.2118/212305-pa","url":null,"abstract":"\u0000 Well placement optimization is one of the most crucial tasks in the petroleum industry. It often involves high risk in the presence of geological uncertainty due to a limited understanding of the subsurface reservoir. Well placement optimization is different from decision selection as countless alternatives are impossible to be enumerated in a decision model (such as the mean-variance model). In many practical applications, the decision criterion of well placement optimization is based on maximizing the risk-adjusted value (mean-variance optimization) to capture different risk attitudes. This approach regards variance as the measure of risk, and it is performed under the expected utility framework. However, investors only dislike the downside volatility below a certain benchmark. The downside-risk approach has been discussed in previous studies, in this paper, it will be introduced in the well placement optimization and discussed under the expected utility framework. It is demonstrated in a synthetic reservoir model with the consideration of spatial heterogeneity, and the comparison between the downside-risk optimization and mean-variance optimization is also presented in this example. The observation implies that well placement optimization is heavily influenced by individuals’ preference to risk. The downside-risk optimization outperforms the mean-variance optimization because it explicitly assesses risk and does not penalize high outcomes.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"1 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82891068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Most oil reservoirs are partially fractured, characterized by finite fracture networks (FFNs) in a sea of isolated fractures. It is necessary to determine size and shape of each FFN explicitly for reservoir simulation. FFN size is correlated with fracture connectivity, which is a function of fracture density, length, and angular scatter. Oil production from FFNs exhibits a long-term dual-porosity behavior. The initial fast rate (Phase I) represents depletion of matrix within FFN, and the subsequent gradual decline phase represents radial flow from the matrix outside the FFN perimeter. Thus, FFN size can be calculated from the cumulative oil production from Phase I, taking into account the pore volume, oil compressibility, and pressure decline. It is not always possible to identify the dual-porosity behavior by visual inspection. A mathematical model is needed to estimate FFN size. For this purpose, a set of three fundamental equations are derived for production rate, cumulative production, and pressure as a function of time. The model is a modified and simplified version of material balance equations with easy analytical solution. It is designed for fractured reservoirs with layer-bound fractures. Production is single-phase black oil under depletion drive. The analytical model was tested on four vertical wells. The unknown parameters such as FFN size, size of well drainage area, and fracture aperture are adjusted until an optimum fit to actual production data is obtained. FFN elliptical shape is estimated from average fracture strike and strike standard deviation. The results are validated by FFN size, fracture length, and aperture measurements from borehole images. The results are approximate but sufficient for preliminary mapping of FFNs with location and size and other critical attributes including fracture drainage area, matrix block size, fracture aperture, and permeability in partially fractured reservoirs.
{"title":"A Simple Analytical Model for Oil Production from Partially Fractured Reservoirs to Estimate Size of Finite Fracture Networks","authors":"S. I. Ozkaya","doi":"10.2118/212296-pa","DOIUrl":"https://doi.org/10.2118/212296-pa","url":null,"abstract":"\u0000 Most oil reservoirs are partially fractured, characterized by finite fracture networks (FFNs) in a sea of isolated fractures. It is necessary to determine size and shape of each FFN explicitly for reservoir simulation. FFN size is correlated with fracture connectivity, which is a function of fracture density, length, and angular scatter. Oil production from FFNs exhibits a long-term dual-porosity behavior. The initial fast rate (Phase I) represents depletion of matrix within FFN, and the subsequent gradual decline phase represents radial flow from the matrix outside the FFN perimeter. Thus, FFN size can be calculated from the cumulative oil production from Phase I, taking into account the pore volume, oil compressibility, and pressure decline. It is not always possible to identify the dual-porosity behavior by visual inspection. A mathematical model is needed to estimate FFN size. For this purpose, a set of three fundamental equations are derived for production rate, cumulative production, and pressure as a function of time. The model is a modified and simplified version of material balance equations with easy analytical solution. It is designed for fractured reservoirs with layer-bound fractures. Production is single-phase black oil under depletion drive.\u0000 The analytical model was tested on four vertical wells. The unknown parameters such as FFN size, size of well drainage area, and fracture aperture are adjusted until an optimum fit to actual production data is obtained. FFN elliptical shape is estimated from average fracture strike and strike standard deviation. The results are validated by FFN size, fracture length, and aperture measurements from borehole images. The results are approximate but sufficient for preliminary mapping of FFNs with location and size and other critical attributes including fracture drainage area, matrix block size, fracture aperture, and permeability in partially fractured reservoirs.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"14 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81791701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Fassihi, A. Alamatsaz, R. Moore, S. Mehta, M. Ursenbach, D. Mallory, P. Pereira Almao, S. C. Gupta, H. Chhina
To understand the role of connate water as a source of hydrogen in oxidation and upgrading of bituminous oil at high temperature, heavy water (D2O) and O-18 enriched water (H2O18) were used as connate water in two different in-situ combustion experiments using a conical tube. Aside from a fundamental understanding of the role of such reactions in in-situ combustion, the results could also potentially help in optimizing in-situ hydrogen generation and upgrading of heavy and bituminous oil. The conical tube had previously been used for understanding the impact of air flux in sustaining the combustion front (please see Alamatsaz et al. 2011, 2012). Significant upgrading was observed in these tests with a produced API gravity of 35 °API compared to the original bitumen gravity of 9 °API. The produced fluids (gases and water) were analyzed with a mass spectrometer to demonstrate the effectiveness of using D2O and H2O18 as tracers in bitumen upgrading reactions and specifically to determine whether the H2O18 would be more effective than D2O in understanding the role of connate water. It was found that heavy water exchanges too readily with hydrogen atoms in the bitumen when the two are mixed. As a result, it could not be reliably used as an indicator for identifying the role of water. The run with O-18 enriched water, however, provided a better indication of water participation in the oxidation reactions. The results obtained from the mass spectrometry analysis of evolved gases and water samples collected in the second test showed that water does split during the reaction and O atoms of indigenous water molecules do combine with C atoms (from the hydrocarbon) to form CO2. These findings open the path for mechanistic considerations around the processes taking place during in-situ combustion and in-situ upgrading reactions.
为了解原生水在烟油高温氧化升级过程中作为氢源的作用，利用锥形管进行了两种不同的原位燃烧实验，分别以重水(D2O)和O-18富集水(H2O18)作为原生水。除了对这些反应在原位燃烧中的作用有了基本的了解外，研究结果还可能有助于优化原位制氢和重质油和沥青油的升级。锥形管以前被用于了解空气通量对维持燃烧前缘的影响(请参见Alamatsaz et al. 2011, 2012)。在这些测试中，与原始沥青的API度9°相比，生产的API度达到35°，效果显著。用质谱仪分析了产出流体(气体和水)，以证明在沥青升级反应中使用D2O和H2O18作为示踪剂的有效性，特别是确定H2O18在理解原生水的作用方面是否比D2O更有效。研究发现，重水与沥青中的氢原子在混合过程中容易发生交换。因此，它不能可靠地用作确定水的作用的指标。然而，用O-18富集的水运行，提供了一个更好的指示水参与氧化反应。从第二次测试中收集的演化气体和水样的质谱分析结果表明，水在反应过程中确实发生了分裂，原生水分子中的O原子确实与(来自碳氢化合物的)C原子结合形成了二氧化碳。这些发现为围绕原位燃烧和原位升级反应过程的机理考虑开辟了道路。
{"title":"A Novel Approach in Understanding the Role of Water in Oxidation and Upgrading Reactions during In-Situ Combustion Oil Recovery, Part B: Interpretations","authors":"M. Fassihi, A. Alamatsaz, R. Moore, S. Mehta, M. Ursenbach, D. Mallory, P. Pereira Almao, S. C. Gupta, H. Chhina","doi":"10.2118/212303-pa","DOIUrl":"https://doi.org/10.2118/212303-pa","url":null,"abstract":"\u0000 To understand the role of connate water as a source of hydrogen in oxidation and upgrading of bituminous oil at high temperature, heavy water (D2O) and O-18 enriched water (H2O18) were used as connate water in two different in-situ combustion experiments using a conical tube. Aside from a fundamental understanding of the role of such reactions in in-situ combustion, the results could also potentially help in optimizing in-situ hydrogen generation and upgrading of heavy and bituminous oil. The conical tube had previously been used for understanding the impact of air flux in sustaining the combustion front (please see Alamatsaz et al. 2011, 2012). Significant upgrading was observed in these tests with a produced API gravity of 35 °API compared to the original bitumen gravity of 9 °API.\u0000 The produced fluids (gases and water) were analyzed with a mass spectrometer to demonstrate the effectiveness of using D2O and H2O18 as tracers in bitumen upgrading reactions and specifically to determine whether the H2O18 would be more effective than D2O in understanding the role of connate water.\u0000 It was found that heavy water exchanges too readily with hydrogen atoms in the bitumen when the two are mixed. As a result, it could not be reliably used as an indicator for identifying the role of water. The run with O-18 enriched water, however, provided a better indication of water participation in the oxidation reactions.\u0000 The results obtained from the mass spectrometry analysis of evolved gases and water samples collected in the second test showed that water does split during the reaction and O atoms of indigenous water molecules do combine with C atoms (from the hydrocarbon) to form CO2. These findings open the path for mechanistic considerations around the processes taking place during in-situ combustion and in-situ upgrading reactions.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"58 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84454352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Polymer flooding is one of the most popular chemical enhanced oil recovery processes in the world, with many large-scale commercial field implementations. It improves mobility ratio and sweep efficiency over waterflooding. However, injection of high volumes of viscous fluid at high injection pressures almost inevitably induces fractures. Induced fractures are not necessarily negative for the flood, and they are documented to improve injectivity and reduce shear degradation of polymer. The big question is on their effect on areal and vertical sweep efficiency. Mangala is one of the largest and most commercially successful polymer floods in the world. The FM-1 unit has the highest stock tank oil initially in place among all the layers of Mangala. Polymer flood sweep efficiency in FM-1 is excellent despite the presence of injection-induced fractures and geological layering. Pressure falloff (PFO) studies are an important and inexpensive source of monitoring the growth of induced fractures in mature polymer floods. Extensive work has been done in the past to derive analytical solutions and numerically assisted solutions for polymer PFOs for non-Newtonian rheology and induced fractures with variable storage. There is also extensive and sometimes contradictory literature on polymer rheology in porous media at both high and low flux rates. Most of the work on polymer PFOs have published a few field examples or synthetic examples and addressed specific issues in depth. We have interpreted 40 PFOs carried out over a period of 1 year. All relevant data sources in the field on geomechanics, polymer rheology in core and rheometers, temperature logs, and saturation logs were integrated to get a consistent estimate for fracture half-length. The results provide insight into the role of induced fractures in improving the sweep efficiency of FM-1. The arguments are backed with literature survey. The work can be used by operators around the world to manage mature polymer floods. The observations mentioned in the work can be used by researchers to fine tune their work. Secondary hypothesis on polymer rheology in porous media is also derived out of the work, which may be of academic interest and merit further work. There are other important aspects of polymer flood PFOs, such as impact of suspended solids and rock dilation near wellbore, which have not been explored in the paper.
{"title":"Decoding the Impact of Injection-Induced Fractures on the Sweep Efficiency of a Mature Polymer Flood through Pressure Falloff Analysis","authors":"Vivek Shankar, Gaurav Sharma","doi":"10.2118/212825-pa","DOIUrl":"https://doi.org/10.2118/212825-pa","url":null,"abstract":"Polymer flooding is one of the most popular chemical enhanced oil recovery processes in the world, with many large-scale commercial field implementations. It improves mobility ratio and sweep efficiency over waterflooding. However, injection of high volumes of viscous fluid at high injection pressures almost inevitably induces fractures. Induced fractures are not necessarily negative for the flood, and they are documented to improve injectivity and reduce shear degradation of polymer. The big question is on their effect on areal and vertical sweep efficiency. Mangala is one of the largest and most commercially successful polymer floods in the world. The FM-1 unit has the highest stock tank oil initially in place among all the layers of Mangala. Polymer flood sweep efficiency in FM-1 is excellent despite the presence of injection-induced fractures and geological layering. Pressure falloff (PFO) studies are an important and inexpensive source of monitoring the growth of induced fractures in mature polymer floods. Extensive work has been done in the past to derive analytical solutions and numerically assisted solutions for polymer PFOs for non-Newtonian rheology and induced fractures with variable storage. There is also extensive and sometimes contradictory literature on polymer rheology in porous media at both high and low flux rates. Most of the work on polymer PFOs have published a few field examples or synthetic examples and addressed specific issues in depth. We have interpreted 40 PFOs carried out over a period of 1 year. All relevant data sources in the field on geomechanics, polymer rheology in core and rheometers, temperature logs, and saturation logs were integrated to get a consistent estimate for fracture half-length. The results provide insight into the role of induced fractures in improving the sweep efficiency of FM-1. The arguments are backed with literature survey. The work can be used by operators around the world to manage mature polymer floods. The observations mentioned in the work can be used by researchers to fine tune their work. Secondary hypothesis on polymer rheology in porous media is also derived out of the work, which may be of academic interest and merit further work. There are other important aspects of polymer flood PFOs, such as impact of suspended solids and rock dilation near wellbore, which have not been explored in the paper.","PeriodicalId":22066,"journal":{"name":"SPE Reservoir Evaluation & Engineering","volume":"49 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80820643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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：