Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.07.005
Shu-Yang Liu , Min-Feng Li , Jia-Yu Chen , Ying Teng , Peng-Fei Wang , Jun-Rong Liu
As one promising CO2 capture, utilization and storage (CCUS) technology, miscible CO2-enhanced oil recovery (CO2-EOR) significantly outperforms immiscible flooding in enhancing oil production and storing CO2. However, achieving CO2 miscible flooding is often hindered by the high minimum miscibility pressure of CO2–oil system in many reservoirs. To address this issue, this study focuses on the mechanisms for enhancing CO2–oil miscibility using different types of surfactants and their blends with ethanol. The effects of fatty alcohol polyoxyethylene ethers (EO), fatty alcohol polyoxypropylene ethers (PO), tributyl citrate (TC), and glyceryl triacetate (GT) on the CO2–oil miscibility pressure are quantitatively analyzed, as well as their synergy with ethanol. Results demonstrated that all tested surfactant additives reduce the CO2–oil miscibility pressure. For ether-based surfactant additives, an increase in the degree of polymerization (CO2-philic groups) weakens the effectiveness to reduce miscibility pressure. Oxygen atoms in the functional group contribute more significantly to miscibility enhancement than carbon atoms. Among ester surfactants, GT achieved the best reduction effect of miscibility pressure (11.82% at 3.0 wt%), attributed to its symmetrical short side-chain structure and ester groups. Furthermore, ethanol exhibited a significant improvement for surfactants in enhancing miscibility. Notably, the reduction of CO2–oil miscibility pressure increases to 27.9% by 3.0 wt% GT blended with 5.0 wt% ethanol. These findings demonstrate that blending surfactants with ethanol is a feasible and effective strategy to facilitate miscible CO2 flooding. This study provides valuable insights and practical guidance for the field implementation of miscible CO2-EOR.
{"title":"Experimental investigation of surfactants and their ethanol blends for CO2–oil miscibility enhancement in CO2-EOR","authors":"Shu-Yang Liu , Min-Feng Li , Jia-Yu Chen , Ying Teng , Peng-Fei Wang , Jun-Rong Liu","doi":"10.1016/j.petsci.2025.07.005","DOIUrl":"10.1016/j.petsci.2025.07.005","url":null,"abstract":"<div><div>As one promising CO<sub>2</sub> capture, utilization and storage (CCUS) technology, miscible CO<sub>2</sub>-enhanced oil recovery (CO<sub>2</sub>-EOR) significantly outperforms immiscible flooding in enhancing oil production and storing CO<sub>2</sub>. However, achieving CO<sub>2</sub> miscible flooding is often hindered by the high minimum miscibility pressure of CO<sub>2</sub>–oil system in many reservoirs. To address this issue, this study focuses on the mechanisms for enhancing CO<sub>2</sub>–oil miscibility using different types of surfactants and their blends with ethanol. The effects of fatty alcohol polyoxyethylene ethers (EO), fatty alcohol polyoxypropylene ethers (PO), tributyl citrate (TC), and glyceryl triacetate (GT) on the CO<sub>2</sub>–oil miscibility pressure are quantitatively analyzed, as well as their synergy with ethanol. Results demonstrated that all tested surfactant additives reduce the CO<sub>2</sub>–oil miscibility pressure. For ether-based surfactant additives, an increase in the degree of polymerization (CO<sub>2</sub>-philic groups) weakens the effectiveness to reduce miscibility pressure. Oxygen atoms in the functional group contribute more significantly to miscibility enhancement than carbon atoms. Among ester surfactants, GT achieved the best reduction effect of miscibility pressure (11.82% at 3.0 wt%), attributed to its symmetrical short side-chain structure and ester groups. Furthermore, ethanol exhibited a significant improvement for surfactants in enhancing miscibility. Notably, the reduction of CO<sub>2</sub>–oil miscibility pressure increases to 27.9% by 3.0 wt% GT blended with 5.0 wt% ethanol. These findings demonstrate that blending surfactants with ethanol is a feasible and effective strategy to facilitate miscible CO<sub>2</sub> flooding. This study provides valuable insights and practical guidance for the field implementation of miscible CO<sub>2</sub>-EOR.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4271-4281"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurately predicting relative permeability is an important issue in the research of multiphase flow in tight reservoirs. Existing predictive models typically rely on the capillary tube bundle model featuring circular cross-sections, often overlooking the impact of pore geometry on fluid flow behavior within reservoirs. In this work, the intermingled fractal theory of porous media is introduced to characterize the intricate local features within the internal space of tight rocks. Initially, iterative rules for diverse fractal units are skillfully designed to capture the actual characteristics of pore cross-sectional shapes. Subsequently, analytical relationships are derived between the iterative parameters and the area, wetted perimeter, and hydraulic diameter of pores generated by these units, followed by the establishment of a relative permeability model that considers pore geometry. The model's validity is confirmed through comparisons with experimental data and published relative permeability models, with correlation coefficients exceeding 0.996. Finally, various factors affecting two-phase flow characteristics are analyzed. The results reveal that pore geometry has a significant impact on flow behavior in porous media. Assuming that the flow channels are cylindrical typically leads to an overestimation of permeability, with the maximum relative error reaching 46.91%. Additionally, the tortuosity fractal dimension is a determinant factor influencing the relative permeability of both wetting and non-wetting fluids, and the phase permeability is sensitive to variations in solid particle size and porosity. The proposed intermingled fractal model enhances the accuracy of evaluating fluid flow characteristics in microscale pore channels and offers a novel framework for simulating porous media with complex geometries.
{"title":"A novel intermingled fractal model for predicting relative permeability in tight oil reservoirs considering microscopic pore geometry","authors":"You Zhou , Song-Tao Wu , Ru-Kai Zhu , Xiao-Hua Jiang , Gan-Lin Hua","doi":"10.1016/j.petsci.2025.06.011","DOIUrl":"10.1016/j.petsci.2025.06.011","url":null,"abstract":"<div><div>Accurately predicting relative permeability is an important issue in the research of multiphase flow in tight reservoirs. Existing predictive models typically rely on the capillary tube bundle model featuring circular cross-sections, often overlooking the impact of pore geometry on fluid flow behavior within reservoirs. In this work, the intermingled fractal theory of porous media is introduced to characterize the intricate local features within the internal space of tight rocks. Initially, iterative rules for diverse fractal units are skillfully designed to capture the actual characteristics of pore cross-sectional shapes. Subsequently, analytical relationships are derived between the iterative parameters and the area, wetted perimeter, and hydraulic diameter of pores generated by these units, followed by the establishment of a relative permeability model that considers pore geometry. The model's validity is confirmed through comparisons with experimental data and published relative permeability models, with correlation coefficients exceeding 0.996. Finally, various factors affecting two-phase flow characteristics are analyzed. The results reveal that pore geometry has a significant impact on flow behavior in porous media. Assuming that the flow channels are cylindrical typically leads to an overestimation of permeability, with the maximum relative error reaching 46.91%. Additionally, the tortuosity fractal dimension is a determinant factor influencing the relative permeability of both wetting and non-wetting fluids, and the phase permeability is sensitive to variations in solid particle size and porosity. The proposed intermingled fractal model enhances the accuracy of evaluating fluid flow characteristics in microscale pore channels and offers a novel framework for simulating porous media with complex geometries.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 3880-3899"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.05.027
Yang Zhang, Hao Yang
Tight sandstone has become an important area in gas exploration. In this study, we propose a 3D seismic reservoir parameter inversion method for tight gas-bearing sandstone reservoirs using dual neural networks. The first network referred to as the inversion network, receives seismic data and predicts reservoir parameters. At well locations, these predictions will be validated based on actual reservoir parameters to evaluate errors. For non-well locations, synthetic seismic data are generated by the application of rock physics forward modeling and seismic reflection coefficient equations. The errors are then calculated by comparing synthetic seismic data with actual seismic data. During the rock physics forward modeling, pseudo reservoir parameters are derived by perturbing the actual reservoir parameters, which are then used to generate pseudo elastic parameters through the modeling. Both the actual and pseudo parameters are then used to train the second network, referred to as the rock physics network. By incorporating the rock physics network, the method effectively alleviates issues such as gradient explosion that may arise from directly integrating rock physics computations into the network, while the inclusion of pseudo parameters enhances the network's generalization capability. The proposed method enables the direct inversion of porosity, clay content, and water saturation from pre-stack seismic data using deep learning, thereby achieving quantitative predictions of reservoir rock physical parameters. The application to the field data from tight sandstone gas reservoirs in southwestern China demonstrates the method has the good capability of indicating the gas-bearing areas and provide high resolution.
{"title":"Direct inversion of 3D seismic reservoir parameters based on dual learning networks","authors":"Yang Zhang, Hao Yang","doi":"10.1016/j.petsci.2025.05.027","DOIUrl":"10.1016/j.petsci.2025.05.027","url":null,"abstract":"<div><div>Tight sandstone has become an important area in gas exploration. In this study, we propose a 3D seismic reservoir parameter inversion method for tight gas-bearing sandstone reservoirs using dual neural networks. The first network referred to as the inversion network, receives seismic data and predicts reservoir parameters. At well locations, these predictions will be validated based on actual reservoir parameters to evaluate errors. For non-well locations, synthetic seismic data are generated by the application of rock physics forward modeling and seismic reflection coefficient equations. The errors are then calculated by comparing synthetic seismic data with actual seismic data. During the rock physics forward modeling, pseudo reservoir parameters are derived by perturbing the actual reservoir parameters, which are then used to generate pseudo elastic parameters through the modeling. Both the actual and pseudo parameters are then used to train the second network, referred to as the rock physics network. By incorporating the rock physics network, the method effectively alleviates issues such as gradient explosion that may arise from directly integrating rock physics computations into the network, while the inclusion of pseudo parameters enhances the network's generalization capability. The proposed method enables the direct inversion of porosity, clay content, and water saturation from pre-stack seismic data using deep learning, thereby achieving quantitative predictions of reservoir rock physical parameters. The application to the field data from tight sandstone gas reservoirs in southwestern China demonstrates the method has the good capability of indicating the gas-bearing areas and provide high resolution.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4037-4051"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.06.016
Yi-Zhong Zhang , Bin Ju , Mao-Lin Zhang , Ping Guo , Jian-Fen Du
The development of gas condensate reservoirs with a large gas cap, thin oil rim, strong bottom water, and natural barriers faces numerous challenges, including reservoir heterogeneity, coning effects, phase changes, and multiphase flow dynamics. The influx of gas and water may lead to a low recovery of the oil rim, while reservoir heterogeneity and natural barriers further exacerbate the uneven distribution of reservoir fluid, complicating development strategies. This paper aims to investigate innovative and effective development strategies for this type of reservoir. A detailed, proportionally scaled numerical simulation is performed based on the experimental results of an artificial sand-filled model, providing novel insights into the dynamic behavior of these reservoirs. By understanding the phase behavior and fluid flow characteristics of the reservoir, the study simulates various strategies for the rational and efficient development of the gas condensate reservoir. These strategies include well patterns and completions, the decision to develop the oil rim or gas cap, depletion rates, the bottom water control, and gas injection. The results show that horizontal wells or highly deviated wells are more suitable for the development of the oil rim, as they provide larger control ranges. The presence of strong bottom water is advantageous for displacement energy supply and pressure maintenance, but it intensifies water coning effects, leading to an earlier breakthrough and a sharp production decline. Therefore, it is preferable to apply highly deviated wells at the oil–gas contact, developing the oil rim at lower rates and smaller pressure gradients, followed by developing the gas cap. This approach can reduce water coning effects and improve recovery, with oil and gas recovery reaching 24.4% and 67.95%, respectively, which is an increase of 16.74% and 17.84% compared to direct depletion development of the gas cap. Due to the strong water bottom, continuous gas injection at the top of the reservoir becomes challenging. This study introduces gas assisted gravity drainage with water control technology, a novel and highly effective approach that addresses the impact of bottom water coning effects on the oil and gas zones and overcomes the limitations of gas flooding in reservoirs with strong bottom water. This method can significantly improve oil and gas recovery, achieving recovery of 39.74% and 84.50%, respectively. Compared to the conventional depletion strategy of sequential oil rim and gas cap development, this method achieves additional improvements of 15.33% and 16.55% in oil and gas recovery, respectively.
{"title":"Development strategies of a gas condensate reservoir with a large gas cap, thin oil rim, strong bottom water, and natural barriers","authors":"Yi-Zhong Zhang , Bin Ju , Mao-Lin Zhang , Ping Guo , Jian-Fen Du","doi":"10.1016/j.petsci.2025.06.016","DOIUrl":"10.1016/j.petsci.2025.06.016","url":null,"abstract":"<div><div>The development of gas condensate reservoirs with a large gas cap, thin oil rim, strong bottom water, and natural barriers faces numerous challenges, including reservoir heterogeneity, coning effects, phase changes, and multiphase flow dynamics. The influx of gas and water may lead to a low recovery of the oil rim, while reservoir heterogeneity and natural barriers further exacerbate the uneven distribution of reservoir fluid, complicating development strategies. This paper aims to investigate innovative and effective development strategies for this type of reservoir. A detailed, proportionally scaled numerical simulation is performed based on the experimental results of an artificial sand-filled model, providing novel insights into the dynamic behavior of these reservoirs. By understanding the phase behavior and fluid flow characteristics of the reservoir, the study simulates various strategies for the rational and efficient development of the gas condensate reservoir. These strategies include well patterns and completions, the decision to develop the oil rim or gas cap, depletion rates, the bottom water control, and gas injection. The results show that horizontal wells or highly deviated wells are more suitable for the development of the oil rim, as they provide larger control ranges. The presence of strong bottom water is advantageous for displacement energy supply and pressure maintenance, but it intensifies water coning effects, leading to an earlier breakthrough and a sharp production decline. Therefore, it is preferable to apply highly deviated wells at the oil–gas contact, developing the oil rim at lower rates and smaller pressure gradients, followed by developing the gas cap. This approach can reduce water coning effects and improve recovery, with oil and gas recovery reaching 24.4% and 67.95%, respectively, which is an increase of 16.74% and 17.84% compared to direct depletion development of the gas cap. Due to the strong water bottom, continuous gas injection at the top of the reservoir becomes challenging. This study introduces gas assisted gravity drainage with water control technology, a novel and highly effective approach that addresses the impact of bottom water coning effects on the oil and gas zones and overcomes the limitations of gas flooding in reservoirs with strong bottom water. This method can significantly improve oil and gas recovery, achieving recovery of 39.74% and 84.50%, respectively. Compared to the conventional depletion strategy of sequential oil rim and gas cap development, this method achieves additional improvements of 15.33% and 16.55% in oil and gas recovery, respectively.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4254-4270"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.07.027
Cun-Fei Ma , Wen-Jun Huang , Jian Zhou , Hong-Zhou Yu , Mei-Yuan Song
The Fengcheng Formation in the Hashan area, located on the northwestern margin of the Junggar Basin, represents a saline-alkaline lake facies with fine-grained mixed sedimentation. This formation is rich in alkaline minerals and serves as a high-quality source rock for hydrocarbon generation in alkaline lakes. However, its lithology is complex, and the origins of the salt minerals remain unclear. This study focuses on the salt minerals in the Fengcheng Formation of the Hashan area. Using core observation, thin section identification, scanning electron microscopy, electron probe micro-analysis, trace and rare earth element analysis, stable isotope analysis, fluid inclusion analysis, and zircon U-Pb dating, the sedimentary age of Fengcheng Formation and the mineralogical and geochemical characteristics of salt minerals were systematically studied. The salt minerals identified in the Fengcheng Formation include calcite, dolomite, eitelite, northupite, shortite, reedmergnerite, and Na-carbonate. According to the different types of salt minerals, the different contact relations between minerals, the different production positions and production styles of mineral combinations, the salt mineral assemblage in the study area is classified into three categories: The combination of calcite, dolomite, shortite, and reedmergnerite, The combination of Na-carbonates, eitelite, shortite, and reedmergnerite, The combination of dolomite, eitelite, shortite, and northupite. Two zircon U-Pb ages, 307.8 ± 2.7 Ma and 308.5 ± 3.5 Ma, span the Carboniferous-Permian boundary, corresponding to an interglacial period within the Late Paleozoic Ice Age, aligning with the development of salt minerals. Salt minerals have the formation modes of sedimentation, replacement and hydrothermal transformation. Terrestrial weathering products, atmospheric, volcanic and hydrothermal processes, residual seawater, clay mineral transformation, thermal evolution of organic matter and tuffaceous alteration are material sources. The salt-forming fluid has the characteristics of weak acid-alkaline, medium-low temperature and high salinity, and is mainly driven by subduction zone high pressure, magmatic heat and gravity. The burial depth, temperature and CO2 concentration required for the formation of salt minerals were clarified, and the evolution sequence of salt-forming fluids from sedimentation to diagenesis and accompanied by hydrothermal (hot water) activities was summarized. The evolution model of salt minerals controlled by different genesis from the first member to the third member of Fengcheng Formation was established. The research findings are significant for understanding the paleoenvironment of the Fengcheng Formation, the formation mechanisms of high-salinity lakes, and the salt formation models.
{"title":"The genetic mechanism of salt minerals in Fengcheng Formation in Hashan area, northwestern margin of Junggar Basin","authors":"Cun-Fei Ma , Wen-Jun Huang , Jian Zhou , Hong-Zhou Yu , Mei-Yuan Song","doi":"10.1016/j.petsci.2025.07.027","DOIUrl":"10.1016/j.petsci.2025.07.027","url":null,"abstract":"<div><div>The Fengcheng Formation in the Hashan area, located on the northwestern margin of the Junggar Basin, represents a saline-alkaline lake facies with fine-grained mixed sedimentation. This formation is rich in alkaline minerals and serves as a high-quality source rock for hydrocarbon generation in alkaline lakes. However, its lithology is complex, and the origins of the salt minerals remain unclear. This study focuses on the salt minerals in the Fengcheng Formation of the Hashan area. Using core observation, thin section identification, scanning electron microscopy, electron probe micro-analysis, trace and rare earth element analysis, stable isotope analysis, fluid inclusion analysis, and zircon U-Pb dating, the sedimentary age of Fengcheng Formation and the mineralogical and geochemical characteristics of salt minerals were systematically studied. The salt minerals identified in the Fengcheng Formation include calcite, dolomite, eitelite, northupite, shortite, reedmergnerite, and Na-carbonate. According to the different types of salt minerals, the different contact relations between minerals, the different production positions and production styles of mineral combinations, the salt mineral assemblage in the study area is classified into three categories: The combination of calcite, dolomite, shortite, and reedmergnerite, The combination of Na-carbonates, eitelite, shortite, and reedmergnerite, The combination of dolomite, eitelite, shortite, and northupite. Two zircon U-Pb ages, 307.8 ± 2.7 Ma and 308.5 ± 3.5 Ma, span the Carboniferous-Permian boundary, corresponding to an interglacial period within the Late Paleozoic Ice Age, aligning with the development of salt minerals. Salt minerals have the formation modes of sedimentation, replacement and hydrothermal transformation. Terrestrial weathering products, atmospheric, volcanic and hydrothermal processes, residual seawater, clay mineral transformation, thermal evolution of organic matter and tuffaceous alteration are material sources. The salt-forming fluid has the characteristics of weak acid-alkaline, medium-low temperature and high salinity, and is mainly driven by subduction zone high pressure, magmatic heat and gravity. The burial depth, temperature and CO<sub>2</sub> concentration required for the formation of salt minerals were clarified, and the evolution sequence of salt-forming fluids from sedimentation to diagenesis and accompanied by hydrothermal (hot water) activities was summarized. The evolution model of salt minerals controlled by different genesis from the first member to the third member of Fengcheng Formation was established. The research findings are significant for understanding the paleoenvironment of the Fengcheng Formation, the formation mechanisms of high-salinity lakes, and the salt formation models.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 3991-4014"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.08.031
Fu-Dong Li , Tian-Yu Chen , Derek Elsworth , Xiao-Jun Yu , Xian-Bao Zheng , Zhi-Guo Wang , Shu-Juan Zhang
Understanding the mechanical behavior and failure characteristics of anisotropic sedimentary rocks under true-triaxial in-situ stress conditions is critical in understanding and mitigating damaging formation slippage in subsurface reservoirs and containment structures. In particular, threshold conditions where structure dominates over intact failure remain undefined. By conducting systematic true-triaxial compression tests, we followed the evolution of deformation and failure in sedimentary rocks across a documented spectrum of lithological and structural characteristics in order to quantify and then classify this cross-impact. The failure features were characterized using acoustic emission (AE) monitoring, optical imaging, X-ray CT scans, and thin-section analysis. We characterized structural and deformational anisotropies in order to define the risk of structurally controlled slip failure. We identified three deformational and failure modes dominated by (I) purely stress-controlled failure, (II) mixed stress–structure-controlled failure, and (III) purely structurally controlled failure. As structural overprinting increased, failure mechanisms were found to shift progressively from Type I to III, thereby progressively capturing inherent rock anisotropy and complex fabric as well as ductile failure. This transition was characterized in terms of two parameters that alternately characterize structural (α) and deformational anisotropies (β) of rocks with these related to key visual, mechanical, and acoustic (AE) indicators. The greater the α (α > 2), the higher the β (β > 0), the more likely the transition from brittle failure to structurally controlled ductile shear reactivation along the bedding.
{"title":"Deformation and failure evolution mechanism of inherently anisotropic sedimentary rocks under true-triaxial stress","authors":"Fu-Dong Li , Tian-Yu Chen , Derek Elsworth , Xiao-Jun Yu , Xian-Bao Zheng , Zhi-Guo Wang , Shu-Juan Zhang","doi":"10.1016/j.petsci.2025.08.031","DOIUrl":"10.1016/j.petsci.2025.08.031","url":null,"abstract":"<div><div>Understanding the mechanical behavior and failure characteristics of anisotropic sedimentary rocks under true-triaxial in-situ stress conditions is critical in understanding and mitigating damaging formation slippage in subsurface reservoirs and containment structures. In particular, threshold conditions where structure dominates over intact failure remain undefined. By conducting systematic true-triaxial compression tests, we followed the evolution of deformation and failure in sedimentary rocks across a documented spectrum of lithological and structural characteristics in order to quantify and then classify this cross-impact. The failure features were characterized using acoustic emission (AE) monitoring, optical imaging, X-ray CT scans, and thin-section analysis. We characterized structural and deformational anisotropies in order to define the risk of structurally controlled slip failure. We identified three deformational and failure modes dominated by (I) purely stress-controlled failure, (II) mixed stress–structure-controlled failure, and (III) purely structurally controlled failure. As structural overprinting increased, failure mechanisms were found to shift progressively from Type I to III, thereby progressively capturing inherent rock anisotropy and complex fabric as well as ductile failure. This transition was characterized in terms of two parameters that alternately characterize structural (<em>α</em>) and deformational anisotropies (<em>β</em>) of rocks with these related to key visual, mechanical, and acoustic (AE) indicators. The greater the <em>α</em> (<em>α</em> > 2), the higher the <em>β</em> (<em>β</em> > 0), the more likely the transition from brittle failure to structurally controlled ductile shear reactivation along the bedding.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4015-4036"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.08.008
Qing-Qing Li , Bo Yu , Jia-Liang Xu , Ning Wang , Shi-Chao Wang , Hui Zhou
The inversion of large sparse matrices poses a major challenge in geophysics, particularly in Bayesian seismic inversion, significantly limiting computational efficiency and practical applicability to large-scale datasets. Existing dimensionality reduction methods have achieved partial success in addressing this issue. However, they remain limited in terms of the achievable degree of dimensionality reduction. An incremental deep dimensionality reduction approach is proposed herein to significantly reduce matrix size and is applied to Bayesian linearized inversion (BLI), a stochastic seismic inversion approach that heavily depends on large sparse matrices inversion. The proposed method first employs a linear transformation based on the discrete cosine transform (DCT) to extract the matrix's essential information and eliminate redundant components, forming the foundation of the dimensionality reduction framework. Subsequently, an innovative iterative DCT-based dimensionality reduction process is applied, where the reduction magnitude is carefully calibrated at each iteration to incrementally reduce dimensionality, thereby effectively eliminating matrix redundancy in depth. This process is referred to as the incremental discrete cosine transform (IDCT). Ultimately, a linear IDCT-based reduction operator is constructed and applied to the kernel matrix inversion in BLI, resulting in a more efficient BLI framework. The proposed method was evaluated through synthetic and field data tests and compared with conventional dimensionality reduction methods. The IDCT approach significantly improves the dimensionality reduction efficiency of the core inversion matrix while preserving inversion accuracy, demonstrating prominent advantages in solving Bayesian inverse problems more efficiently.
{"title":"Incremental dimensionality reduction for efficiently solving Bayesian inverse problems","authors":"Qing-Qing Li , Bo Yu , Jia-Liang Xu , Ning Wang , Shi-Chao Wang , Hui Zhou","doi":"10.1016/j.petsci.2025.08.008","DOIUrl":"10.1016/j.petsci.2025.08.008","url":null,"abstract":"<div><div>The inversion of large sparse matrices poses a major challenge in geophysics, particularly in Bayesian seismic inversion, significantly limiting computational efficiency and practical applicability to large-scale datasets. Existing dimensionality reduction methods have achieved partial success in addressing this issue. However, they remain limited in terms of the achievable degree of dimensionality reduction. An incremental deep dimensionality reduction approach is proposed herein to significantly reduce matrix size and is applied to Bayesian linearized inversion (BLI), a stochastic seismic inversion approach that heavily depends on large sparse matrices inversion. The proposed method first employs a linear transformation based on the discrete cosine transform (DCT) to extract the matrix's essential information and eliminate redundant components, forming the foundation of the dimensionality reduction framework. Subsequently, an innovative iterative DCT-based dimensionality reduction process is applied, where the reduction magnitude is carefully calibrated at each iteration to incrementally reduce dimensionality, thereby effectively eliminating matrix redundancy in depth. This process is referred to as the incremental discrete cosine transform (IDCT). Ultimately, a linear IDCT-based reduction operator is constructed and applied to the kernel matrix inversion in BLI, resulting in a more efficient BLI framework. The proposed method was evaluated through synthetic and field data tests and compared with conventional dimensionality reduction methods. The IDCT approach significantly improves the dimensionality reduction efficiency of the core inversion matrix while preserving inversion accuracy, demonstrating prominent advantages in solving Bayesian inverse problems more efficiently.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4102-4116"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.09.014
Heng-Rui Zhang , Yi-Nao Su , Mao-Lin Liao , Hong-Yu Wu , Hai-Yan Zhang , Hao-Yu Wang , Ke Liu
The exploitation of oil resources has now extended to ultra-deep formations, with depths even exceeding 10,000 m. During drilling operations, the bottomhole temperature (BHT) can surpass 240 °C. Under such high-temperature conditions, measurement while drilling (MWD) instruments are highly likely to malfunction due to the inadequate temperature resistance of their electronic components. As a wellbore temperature control approach, the application of thermal insulated drill pipe (TIDP) has been proposed to manage the wellbore temperature in ultra-deep wells. This paper developed a temperature field model for ultra-deep wells by coupling the interactions of multiple factors on the wellbore temperature. For the first time, five distinct TIDP deployment methods were proposed, and their corresponding wellbore temperature variation characteristics were investigated, and the heat transfer laws of the ultra-deep wellbore-formation system were quantitatively elucidated. The results revealed that TIDP can effectively restrain the rapid rise in the temperature of the drilling fluid inside the drill string by reducing the heat flux of the drill string. Among the five deployment methods, the method of deploying TIDP from the bottomhole upwards exhibits the best performance. For a 12,000 m simulated well, when 6000 m of TIDP are deployed from the bottomhole upwards, the BHT decreases by 52 °C, while the outlet temperature increases by merely 1 °C. This not only achieves the objective of wellbore temperature control but also keeps the temperature of the drilling fluid at the outlet of annulus at a relatively low level, thereby reducing the requirements for the heat exchange equipment on the ground. The novel findings of this study provide significant guidance for wellbore temperature control in ultra-deep and ultra-high-temperature wells.
{"title":"Effects of different thermal insulated drill pipe deployment methods on wellbore temperature control in ultra-deep wells","authors":"Heng-Rui Zhang , Yi-Nao Su , Mao-Lin Liao , Hong-Yu Wu , Hai-Yan Zhang , Hao-Yu Wang , Ke Liu","doi":"10.1016/j.petsci.2025.09.014","DOIUrl":"10.1016/j.petsci.2025.09.014","url":null,"abstract":"<div><div>The exploitation of oil resources has now extended to ultra-deep formations, with depths even exceeding 10,000 m. During drilling operations, the bottomhole temperature (BHT) can surpass 240 °C. Under such high-temperature conditions, measurement while drilling (MWD) instruments are highly likely to malfunction due to the inadequate temperature resistance of their electronic components. As a wellbore temperature control approach, the application of thermal insulated drill pipe (TIDP) has been proposed to manage the wellbore temperature in ultra-deep wells. This paper developed a temperature field model for ultra-deep wells by coupling the interactions of multiple factors on the wellbore temperature. For the first time, five distinct TIDP deployment methods were proposed, and their corresponding wellbore temperature variation characteristics were investigated, and the heat transfer laws of the ultra-deep wellbore-formation system were quantitatively elucidated. The results revealed that TIDP can effectively restrain the rapid rise in the temperature of the drilling fluid inside the drill string by reducing the heat flux of the drill string. Among the five deployment methods, the method of deploying TIDP from the bottomhole upwards exhibits the best performance. For a 12,000 m simulated well, when 6000 m of TIDP are deployed from the bottomhole upwards, the BHT decreases by 52 °C, while the outlet temperature increases by merely 1 °C. This not only achieves the objective of wellbore temperature control but also keeps the temperature of the drilling fluid at the outlet of annulus at a relatively low level, thereby reducing the requirements for the heat exchange equipment on the ground. The novel findings of this study provide significant guidance for wellbore temperature control in ultra-deep and ultra-high-temperature wells.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4174-4194"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.petsci.2025.06.007
Hai-Yang Chen , Liang Xue , Li Liu , Gao-Feng Zou , Jiang-Xia Han , Yu-Bin Dong , Meng-Ze Cong , Yue-Tian Liu , Seyed Mojtaba Hosseini-Nasab
With the rapid development of deep learning neural networks, new solutions have emerged for addressing fluid flow problems in porous media. Combining data-driven approaches with physical constraints has become a hot research direction, with physics-informed neural networks (PINNs) being the most popular hybrid model. PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements, fast training speeds, strong generalization capabilities, and broad applicability. Despite success in homogeneous settings, standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells. This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir. To address these challenges, this study proposes a physics-informed graph neural network (PIGNN) model. The PIGNN model treats the entire field as a whole, integrating information from neighboring grids and physical laws into the solution for the target grid, thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids. The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir, achieving an average error and R2 score of 6.710 × 10−4 and 0.998, respectively, which confirms the effectiveness of model. Compared to the conventional PINN model, the average error was reduced by 76.93%, the average R2 score increased by 3.56%. Moreover, evaluating robustness, training the PIGNN model using only 54% and 76% of the original data yielded average relative error reductions of 58.63% and 56.22%, respectively, compared to the PINN model. These results confirm the superior performance of this approach compared to PINN.
{"title":"Physics-informed graph neural network for predicting fluid flow in porous media","authors":"Hai-Yang Chen , Liang Xue , Li Liu , Gao-Feng Zou , Jiang-Xia Han , Yu-Bin Dong , Meng-Ze Cong , Yue-Tian Liu , Seyed Mojtaba Hosseini-Nasab","doi":"10.1016/j.petsci.2025.06.007","DOIUrl":"10.1016/j.petsci.2025.06.007","url":null,"abstract":"<div><div>With the rapid development of deep learning neural networks, new solutions have emerged for addressing fluid flow problems in porous media. Combining data-driven approaches with physical constraints has become a hot research direction, with physics-informed neural networks (PINNs) being the most popular hybrid model. PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements, fast training speeds, strong generalization capabilities, and broad applicability. Despite success in homogeneous settings, standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells. This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir. To address these challenges, this study proposes a physics-informed graph neural network (PIGNN) model. The PIGNN model treats the entire field as a whole, integrating information from neighboring grids and physical laws into the solution for the target grid, thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids. The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir, achieving an average <span><math><mrow><msub><mi>L</mi><mn>2</mn></msub></mrow></math></span> error and <em>R</em><sup>2</sup> score of 6.710 × 10<sup>−4</sup> and 0.998, respectively, which confirms the effectiveness of model. Compared to the conventional PINN model, the average <span><math><mrow><msub><mi>L</mi><mn>2</mn></msub></mrow></math></span> error was reduced by 76.93%, the average <em>R</em><sup>2</sup> score increased by 3.56%. Moreover, evaluating robustness, training the PIGNN model using only 54% and 76% of the original data yielded average relative <span><math><mrow><msub><mi>L</mi><mn>2</mn></msub></mrow></math></span> error reductions of 58.63% and 56.22%, respectively, compared to the PINN model. These results confirm the superior performance of this approach compared to PINN.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 4240-4253"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 paper, experiments were carried out to investigate the retention of liquid hydrocarbons in kerogen type I. The study focuses on the mudstone from the Lucaogou Formation in the Junggar Basin of China. To prepare samples of kerogen with varying degrees of maturity, artificial pyrolysis was used. Swelling experiments with three different types of liquid hydrocarbons were then conducted. The results revealed that the peak swelling adsorption capacity of type I kerogen for liquid hydrocarbons occurred at Easy%Ro = 1.07. Additionally, the kerogen showed a selective ability to retain aromatic hydrocarbons throughout the entire process compared to alkane. The order of hydrocarbon expulsion from source rocks was established as follows: short-chain alkanes > cycloalkanes/long-chain alkanes > aromatics with alkyl groups > polycyclic aromatic hydrocarbons. This study also developed a model for evaluating the swelling capacity of kerogen. This model was capable of evaluating the total swelling state of liquid hydrocarbons without considering the adsorption state, which was not possible in previous experimental work. According to this model, the swelling ability of long-chain alkanes and polycyclic aromatic hydrocarbons in type I kerogen was high, while the swelling ability of cycloalkanes was weak, and most of them existed in the form of adsorption. This study suggests that paraffin and asphaltenes may affect the expulsion of shale oil and heavy oil in the form of swelling state, particularly in immature source rocks. This finding provides a new direction for research on hydrocarbon source rock evaluation and unconventional oil exploration.
{"title":"Study on swelling and retention of liquid hydrocarbon compounds by type I kerogen","authors":"Tian Liang , Yan-Rong Zou , Zha-Wen Zhan , Ping-An Peng","doi":"10.1016/j.petsci.2025.07.013","DOIUrl":"10.1016/j.petsci.2025.07.013","url":null,"abstract":"<div><div>In this paper, experiments were carried out to investigate the retention of liquid hydrocarbons in kerogen type I. The study focuses on the mudstone from the Lucaogou Formation in the Junggar Basin of China. To prepare samples of kerogen with varying degrees of maturity, artificial pyrolysis was used. Swelling experiments with three different types of liquid hydrocarbons were then conducted. The results revealed that the peak swelling adsorption capacity of type I kerogen for liquid hydrocarbons occurred at Easy%<em>R</em><sub>o</sub> = 1.07. Additionally, the kerogen showed a selective ability to retain aromatic hydrocarbons throughout the entire process compared to alkane. The order of hydrocarbon expulsion from source rocks was established as follows: short-chain alkanes > cycloalkanes/long-chain alkanes > aromatics with alkyl groups > polycyclic aromatic hydrocarbons. This study also developed a model for evaluating the swelling capacity of kerogen. This model was capable of evaluating the total swelling state of liquid hydrocarbons without considering the adsorption state, which was not possible in previous experimental work. According to this model, the swelling ability of long-chain alkanes and polycyclic aromatic hydrocarbons in type I kerogen was high, while the swelling ability of cycloalkanes was weak, and most of them existed in the form of adsorption. This study suggests that paraffin and asphaltenes may affect the expulsion of shale oil and heavy oil in the form of swelling state, particularly in immature source rocks. This finding provides a new direction for research on hydrocarbon source rock evaluation and unconventional oil exploration.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"22 10","pages":"Pages 3960-3966"},"PeriodicalIF":6.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145384416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号