Petroleum Science最新文献

{"title":"Potential origin of natural vanadium hydride in the Earth's shallow mantle","authors":"Yu-Tong Su,&nbsp;Zhi-Jun Jin,&nbsp;Ren-Biao Tao,&nbsp;Jin-Tao Zhu,&nbsp;Run-Chao Liu,&nbsp;Lu Wang,&nbsp;Hao-Zhe Zhang,&nbsp;Shubham Choudhary","doi":"10.1016/j.petsci.2026.04.006","DOIUrl":"10.1016/j.petsci.2026.04.006","url":null,"abstract":"<div><div>Metal hydrides have emerged as promising candidates for energy storage in clean hydrogen systems. While these materials are typically synthesized under controlled laboratory conditions, naturally occurring vanadium hydride (VH₂) has been identified in mantle-derived igneous rocks, suggesting that it may from in geological environments. In this study, we investigate the formation mechanisms and melting behavior of vanadium hydride under mantle conditions by integrating high-pressure experimental techniques with molecular dynamics simulations. Our experimental results demonstrate that hydrous minerals, such as Mg(OH)₂, can react with elemental vanadium (V₀) at elevated temperatures (1000–1500 °C) and pressures (0.8–3 GPa) to form vanadium hydride. Molecular dynamics simulations further reveal that the melting temperature of VH₂ at 1 GPa ranges between 1400 and 1500 °C. These findings suggest that natural vanadium hydride may originate in the Earth's shallow mantle and subsequently migrate to the crust or surface through magmatic processes. This study provides a theoretical basis for further exploration of naturally occurring metal hydrides and supports the development of synthetic metal hydrides for solid-state hydrogen storage.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5984-5991"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Solubility and dissolution mechanism of novel multi-ester headgroup surfactants in supercritical CO2","authors":"Ning Xu,&nbsp;Yan-Ling Wang,&nbsp;Baojun Bai,&nbsp;Shi-Zhang Cui,&nbsp;Yu Zhang,&nbsp;Wen-Jing Shi,&nbsp;Zhao-Nian Zhang,&nbsp;Wen-Hui Ding,&nbsp;Pei-Xu Ma,&nbsp;Zan Gao","doi":"10.1016/j.petsci.2026.06.010","DOIUrl":"10.1016/j.petsci.2026.06.010","url":null,"abstract":"<div><div>To address the limited solubility and applicability of conventional hydrocarbon surfactants in supercritical CO₂, a series of multi-ester headgroup surfactants were designed and synthesized by leveraging the CO₂-philic properties of ester groups. The molecular structures were characterized using Fourier transform infrared (FT-IR) spectroscopy and ¹H NMR. A custom-designed laser-based apparatus was developed to quantify surfactant solubility and systematically investigate phase behavior in CO₂. Molecular dynamics (MD) simulations were employed to elucidate structure–solubility relationships across multiple scales, including solubility parameters, interaction energies, radial distribution functions (RDFs), and free volume fractions. Results indicate that, at 323.15 K, the cloud-point pressure of a 1 wt% multi-ester headgroup surfactant in CO₂ is below 12 MPa. This value is substantially lower than the typical minimum miscibility pressure (MMP) between crude oil and CO₂ in Chinese reservoirs. The number and arrangement of ester groups synergistically affect solubility, with linear configurations outperforming cyclic ones and an optimal ester count (<em>n</em> = 5) maximizing solubility. Surfactant–CO₂ interactions are primarily governed by van der Waals and electrostatic forces, with van der Waals contributions exceeding 70%. RDF analysis reveals that carbonyl oxygen (O1) and terminal methyl carbon (C1) in the ester group are the primary CO₂ binding sites, enhancing CO₂-philicity via Lewis acid–base (LA–LB) and dispersion interactions. Linear ester side chains exhibit enhanced flexibility and conformational adaptability. The free volume fraction initially increases and then decreases with increasing ester group count, with the optimal configuration (MEG-L5) showing maximal chain extension and CO₂ contact efficiency during dissolution. Given that hydrocarbon surfactants primarily interact with CO₂ through LA–LB and dispersion forces, the design focus is on optimizing the density and arrangement of CO₂-philic segments. This modulates entropy changes to enhance favorable enthalpic interactions. This study elucidates the dissolution mechanisms of multi-ester headgroup surfactants in supercritical CO₂ (scCO₂) and identifies the molecular factors controlling their CO₂-philicity. These insights provide a theoretical basis for designing green, fluorine-free, cost-effective CO₂-philic surfactants for enhanced oil recovery.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5800-5818"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Influence of pore structure on high-voltage electrical pulse rock breaking: A coupled numerical approach","authors":"Yu-Wei Li,&nbsp;Xin-Hui Guo,&nbsp;Xue-Wei Liu,&nbsp;Yuan Pan,&nbsp;Yong-Hui Xiao,&nbsp;Hui-Min Tang,&nbsp;Dmitriy A. Martyushev,&nbsp;Zi-Han Sun","doi":"10.1016/j.petsci.2026.05.032","DOIUrl":"10.1016/j.petsci.2026.05.032","url":null,"abstract":"<div><div>High-voltage electrical pulse (HVEP) rock-breaking technology is extensively utilized due to its advantages of high energy density, controllable fragmentation zones, and tunable energy output. However, the intricate pore structures within rock masses can significantly affect their electrical and mechanical properties, thereby influencing HVEP performance. To elucidate the mechanisms by which pore structure influences rock-breaking outcomes, this study integrates the finite difference method (FDM) and stochastic propagation model to simulate the electrical breakdown paths. Furthermore, the numerical simulations of HVEP-induced rock fragmentation are conducted using the discrete element method (DEM) based on thermo-mechanical coupling theory. The results demonstrate that the pore structure strongly impacts both the complexity of the breakdown path and the resulting fragmentation efficiency. In non-porous rocks, breakdown paths are primarily governed by rock heterogeneity, resulting in a complex branching structure. In contrast, in porous rocks, breakdown paths are guided by pore spaces, which reduces path complexity. Notably, under constant porosity, pore diameter exerts a nonlinear influence on path complexity. Smaller pore diameters, which correspond to a higher pore density, significantly enhance path complexity. Conversely, as pore diameter increases and the number of pores decreases, path complexity exhibits a ‘rebound’ as the governing mechanism shifts back to inherent rock heterogeneity. Moreover, the pore structure induces stress concentration, promoting fracture initiation and propagation, which enhances rock-breaking performance. Although rocks with small pores exhibit higher breakdown-path complexity, their fragmentation efficiency is lower than that of large-pore rocks. This finding highlights the dominant role of pore-induced stress field alterations. This study provides valuable theoretical guidance for the application of HVEP rock-breaking technology in porous rock fragmentation.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5853-5869"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of initial damage degrees on crack propagation and permeability characteristics of coal subjected to cyclic loading","authors":"Qi-Can Ran,&nbsp;Yun-Pei Liang,&nbsp;Chun-Feng Ye,&nbsp;Teng-Fei Ma,&nbsp;Zi-Han Chen","doi":"10.1016/j.petsci.2026.06.011","DOIUrl":"10.1016/j.petsci.2026.06.011","url":null,"abstract":"<div><div>Once deep coal rock is disturbed by engineering operations, coal exhibits different initial damage degrees (IDDs), which may govern subsequent crack propagation and permeability evolution under cyclic stress. Therefore, cyclic loading and permeability tests, combined with acoustic emission (AE) and computed tomography (CT) scanning techniques, were conducted to systematically investigate the effects of the IDD on the crack propagation and permeability evolution of coal. The results indicated that high IDDs led to a significant decrease in the peak strength and deformation modulus of the coal samples and a significant increase in the dissipated energy percentage. Furthermore, AE analyses showed that the <em>b</em> value dropped from 1.863 to 0.784 and the energy critical index decreased, whereas the multifractal spectrum width and subset parameter increased, indicating a shift from small-scale microcracking to more complex, large-scale crack coalescence in coal specimens with higher IDDs. Permeability generally decreased with increasing stress, but coal specimens with high IDDs exhibited a transient permeability recovery at high stress levels and a larger post-failure permeability enhancement, reflecting the competition between compaction and crack propagation. Moreover, CT-based analysis shows that crack volume percentage increased from 2.821% to 8.968% and fractal dimension rose from 2.011 to 2.052 with increasing IDDs, correlating well with permeability evolution. These results provide useful implications for stability and permeability characteristics in coal reservoirs and deep geo-energy engineering.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5738-5756"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Seismic resolution enhancement via deep learning with teacher-student network and domain adaptation","authors":"Han-Peng Cai,&nbsp;Hao-Nan Zhang,&nbsp;Li-Yu Zhang,&nbsp;Suo Cheng","doi":"10.1016/j.petsci.2026.04.042","DOIUrl":"10.1016/j.petsci.2026.04.042","url":null,"abstract":"<div><div>High-resolution processing of seismic signals is crucial for enhancing the ability to characterize underground geological structures in detail and improving the accuracy of thin-layer reservoir identification. Traditional seismic high-resolution algorithms suffer from issues such as poor robustness, low computational efficiency, and neglect of inter-channel structural relationships. Meanwhile, most mainstream deep learning-based high-resolution methods rely on end-to-end networks, lacking guidance from prior information and ignoring differences between data domains, resulting in insufficient generalization capabilities. Therefore, this paper proposes a domain-adaptive knowledge distillation deep learning network for seismic data, DAKD-Net (Domain-Adaptive Knowledge Distillation Network). This method establishes a teacher-student network model using forward simulation datasets as training samples. The teacher network module establishes physical constraints between low- and high-resolution data, extracting high-frequency prior information during the guidance phase to direct the student network module in restoring details without prior conditions. Domain adaptation then generalizes the model to real seismic data, enhancing its generalization capability and structural representation accuracy in actual work area data. Structurally, DAKD-Net adopts a U-net backbone to fully extract spatial structural information across multi-track seismic profiles. Its training mechanism enables prior knowledge transfer through the teacher-student network, allowing high-resolution data recovery without prior information. In application, domain adaptation fine-tuning strategies enhance the network's generalization capability and structural representation effectiveness in real-world work areas. Experimental results demonstrate that the proposed method outperforms traditional approaches and classical deep networks in both vertical resolution and complex structural detail recovery, exhibiting robust performance and practical applicability.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5433-5449"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optical and rheological properties of carbon quantum dots regenerated aged asphalt based on molecular simulation and experiments","authors":"Xin-Xing Zhou,&nbsp;Tian-Qi Zong,&nbsp;Ling-Lin Li,&nbsp;Seung-Soo Kim","doi":"10.1016/j.petsci.2026.06.048","DOIUrl":"10.1016/j.petsci.2026.06.048","url":null,"abstract":"<div><div>To investigate the rheological and optical properties of carbon quantum dots (CQDs)-regenerated aged asphalt (CQDsRA), four types of CQDs, C₃₄H₁₄O₂ (CQDs-1), C₅₂H₇₂O₁₄ (CQDs-2), C₃₉H₃₀N₆O₁₀ (CQDs-3), and C₅₂H₇₂O₁₀S₄ (CQDs-4), were employed to regenerate aged base asphalt using first-principles calculations and molecular dynamics simulations. The optical properties of both the CQDs and the resulting CQDsRA, including band structure, electron density, molecular orbitals, and optical responses, were systematically examined. The results reveal that the type of CQDs significantly influences the optical response of CQDsRA. Moreover, the different CQDs also markedly affect the fundamental properties and regeneration efficacy of the modified asphalt. Among the four types, CQDs-4 exhibits the strongest fluorescence intensity, whereas CQDs-1 shows the weakest. The molecular trajectories of the CQDs resemble those of a standard pyrometric cone during the regeneration process. Additionally, the chemical reactions occur between the CQDs and the aged asphalt, leading to the formation of a cross-linked network structure.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5960-5971"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In situ-generated silica nanonetworks for energy-efficient enhanced oil recovery: Breaking the migration–aggregation dilemma in heterogeneous reservoirs","authors":"Yi-An Zhao,&nbsp;Xun Qi,&nbsp;Su-Chen Xiao,&nbsp;Zhang Luo,&nbsp;Yun-Long Liu,&nbsp;Tian-Jiang Wu,&nbsp;Jie Wen,&nbsp;Cai Chen,&nbsp;Hui Zhang","doi":"10.1016/j.petsci.2026.05.025","DOIUrl":"10.1016/j.petsci.2026.05.025","url":null,"abstract":"<div><div>Maximizing energy efficiency in hydrocarbon extraction remains hampered by water channeling in heterogeneous reservoirs, where conventional pre-synthesized nanoparticles often fail due to the unresolved migration–aggregation dilemma. We report a novel strategy that addresses the long-standing migration–aggregation dilemma of pre-synthesized nanoparticles by enabling the <em>in situ</em> formation of functional silica nanonetworks directly within reservoir pore throats. Our approach employs a tetraethyl orthosilicate (TEOS) emulsion that, upon controlled demulsification, undergoes hydrolysis and condensation to generate silica nanoparticles <em>in situ</em>. Glycerol mediates dielectric tuning and modulates nanoparticle zeta potential, allowing precise control over particle size and distribution to match medium-to high-permeability zones (100–1000 mD). Displacement experiments demonstrate exceptional performance, achieving a water shutoff efficiency of over 90% while retaining more than 70% of oil-phase permeability. Mechanistic studies reveal that surfactants released during demulsification play a dual role. They enhance silica nucleation rate by reducing the interfacial energy barriers and adsorb onto nanoparticle surfaces, conferring strong hydrophobicity for selective water shutoff. Furthermore, the <em>in situ</em>-generated SiO₂ nanoparticles form a stable three-dimensional plugging network through physical bridging, aggregation, and surfactant-mediated anchoring on rock surfaces. Compared to conventional high-viscosity systems, the ultra-low viscosity and <em>in situ</em> formation strategy of this system results in lower operational energy consumption. This strategy improves nanoparticle delivery efficiency and provides an energy-efficient approach for enhanced oil recovery.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5770-5787"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigation on force responses in ridged diamond element breaking granite","authors":"Chao Xiong,&nbsp;Xin-Long Li,&nbsp;Sen-Lin He,&nbsp;Zhong-Wei Huang,&nbsp;Huai-Zhong Shi,&nbsp;Wen-Hao He","doi":"10.1016/j.petsci.2026.05.026","DOIUrl":"10.1016/j.petsci.2026.05.026","url":null,"abstract":"<div><div>Ridged diamond element (RDE), a new 3D shaped PDC cutter, has higher impact and wear resistance than conventional plane PDC cutter. Furthermore, RDE also can significantly improve the rock-breaking efficiency of PDC drill bits in hard rock. The investigation on force response during the RDE breaking rock is a critically important work for revealing the rock breakage mechanism and designing the AxeBlade PDC bit. However, to our best knowledge, there is still a lack of systematic and in-depth research on the force response of RDE cutting hard rock. This paper conducted a series of single cutter tests on granite, and measured the triaxial forces of RDE under different depth of cut (DOC) and back rake angle. Then, a mechanical model of interaction between RDE and rock was established through theoretical derivation. The model accuracy is validated by experimental data, and the average error of prediction cutting force and normal force is 3.13% and 3.68% respectively. Lastly, the influence of DOC, back rake angle, cutter diameter and ridge angle on cutting force and normal force was analyzed, and the influence weights of four parameters were revealed by random forest method. The results show that the cutting force and the normal force increase with DOC in a quadratic polynomial relationship. With the increase in back rake angle, cutter diameter and ridge angle, the cutting force and normal force show an increasing trend, but there are significant differences in the rate of increase. When the back rake angle increases from 10° to 30° for DOC of 2 mm, the cutting force and normal force increases by 416.22 and 1008.04 N respectively. The cutter diameter increases from 10 to 20 mm, the cutting force and normal force increases by 1318.13 and 1739.10 N respectively. The ridge angle increases from 100° to 165°, the cutting force and normal force increases by 1276.03 and 750.87 N respectively. The DOC has the largest influence weight on the force response of RDE in four parameters. The findings of this work can provide theoretical guidance for the cutter arrangement of AxeBlade PDC bit.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5610-5623"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impacts of temperature and pressure on light hydrocarbon parameters during gas washing: Results from simulation experiments","authors":"Wen-Na Liu,&nbsp;Wang-Lu Jia,&nbsp;Qiang Wang,&nbsp;Jian Chen,&nbsp;Jin-Bu Li,&nbsp;Ping-An Peng","doi":"10.1016/j.petsci.2026.03.031","DOIUrl":"10.1016/j.petsci.2026.03.031","url":null,"abstract":"<div><div>As a secondary alteration process in oil reservoirs and a critical mechanism for light oil and condensate formation, gas washing is predominantly governed by temperature and pressure. However, the dynamic effects of these two factors on the light hydrocarbon composition of residual oil remain unclear. This study developed a methane-pressurized gas washing simulation apparatus that can maintain predetermined pressure and temperature conditions throughout the experiment. The aim of this study is to investigate the variation patterns of light hydrocarbons under different temperature and pressure conditions, and to further evaluate their potential influences in identifying the type and depositional environments of source organic matter, quantifying thermal maturity, and identifying secondary alteration processes. By establishing the relationship between the remaining mass fraction of light hydrocarbons and time, the loss rate constants for individual components were obtained. Comparative analysis revealed that the loss rates of light hydrocarbons increased significantly with rising temperature during gas washing, but were suppressed with increasing pressure. Among the parameters indicative of source rock type and depositional environment, the Mango parameter <em>K</em>₁ and the C₇ ternary diagram were less affected by gas washing and changes in temperature and pressure. In contrast, parameter <em>K</em>₂ and Halpern's oil-gas correlation parameters (C₇-OCSD star diagram) underwent considerable alterations. Gas washing resulted in an increase in both MCH/(MCH + <em>n</em>C₇) and Tol/(Tol + MCH) ratios, a trend opposite to that induced by increasing maturity. Furthermore, elevated temperature significantly promoted an increase in the MCH/(MCH + <em>n</em>C₇) ratio, whereas increased pressure exerted an inhibitory effect on this ratio. The variation in secondary alteration parameters (Tol/<em>n</em>C₇ and <em>n</em>C₇/MCH) is consistent with classical theory: gas washing leads to an increase in Tol/<em>n</em>C₇ and a decrease in <em>n</em>C₇/MCH. Furthermore, higher temperatures result in a smaller increase in Tol/<em>n</em>C₇, while greater pressure leads to a more pronounced increase. These findings enhance the understanding of how temperature and pressure govern gas washing mechanisms, providing critical insights for more accurately identifying hydrocarbon origin, maturity, and secondary alteration processes in complex geological contexts.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5509-5523"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stochastic simulation and finely characterization of heterogeneous reservoirs based on MSGAN and BicycleGAN","authors":"Xue-Chao Wu,&nbsp;Wen-Yao Fan,&nbsp;Shi-Jie Peng,&nbsp;Xin-Hua Xu,&nbsp;Feng Deng,&nbsp;Xi-Xi Du","doi":"10.1016/j.petsci.2026.07.009","DOIUrl":"10.1016/j.petsci.2026.07.009","url":null,"abstract":"<div><div>Subsurface reservoirs can be constructed based on Generative Adversarial Networks (GANs), but some challenges cannot be overlooked. For example, both prior geological patterns and knowledge are hardly quantified so that GANs cannot be trained without sufficient samples. Meanwhile, the heterogeneous patterns are difficult to characterize, with less consideration of multiple geological constraints, resulting in poor modeling accuracy and reliability. Therefore, we integrated Multi-Stage GAN (MSGAN) and BicycleGAN to finely characterize heterogeneous reservoirs. Specifically, MSGAN can extract and quantify geological patterns based on multiscale representations of pyramid structure. Through concurrent training and parameter inheritance, these simulated prior samples share the same feature space with the Training Image (TI) so that they can support BicycleGAN's training. The bijective consistency between latent codes and output modes is established, and reservoir characterization is achieved through the nonlinear mapping between multiple conditioning data and output modes. Both synthetic cases and field application were introduced to verify the applicability of the proposed modeling framework. During data augmentation based on MSGAN, different simulations preserve a better generation diversity, and the complex geological pattern fitting with multiple geological constraints can be reproduced by BicycleGAN. Each conditional simulation is similar to TI in terms of spatial variability, channel connectivity and spatial structures, significantly reducing modeling uncertainties.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5334-5358"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}