Petroleum Science最新文献

{"title":"In-situ anaerobic microbial oxidation of alkane gases within tight shale reservoir in the Ordos Basin: Evidence from intramolecular isotope compositions of desorbed propane","authors":"Rui-Liang Guo,&nbsp;Juske Horita,&nbsp;Ji-Hong Li,&nbsp;Yun-Chao Hou,&nbsp;Jun-Lin Chen,&nbsp;Peng Liu","doi":"10.1016/j.petsci.2026.03.052","DOIUrl":"10.1016/j.petsci.2026.03.052","url":null,"abstract":"<div><div>Anaerobic microbial oxidation of hydrocarbons is a key biogeochemical process influencing hydrocarbon compositions and the carbon cycle in subsurface systems. In this study, the microbial oxidation of propane within the Chang 7 shale of the Ordos Basin was investigated through intramolecular carbon isotope analysis of thermally desorbed gases. The results show that samples affected by microbial activity exhibit elevated <em>Δ</em>_C-T values (3.9‰–7.9‰), <em>δ</em>¹³C_central values ranging from −27.2‰ to −22.9‰, and <em>δ</em>¹³C_terminal values from −31.1‰ to −30.5‰, significantly higher than non-degraded samples. These isotopic characteristics indicate that microorganisms preferentially oxidize the central carbon atom in propane, while the terminal carbon retains its original isotopic signature. The distinct position-specific isotopic patterns serve as sensitive indicators of microbial degradation in shale reservoirs. Furthermore, sedimentological data suggest that microbial degradation potential is closely linked to the sand supply from different sedimentary facies. Toward deeper lacustrine settings, reductions in sandy interlayer thickness and grain size, along with enhanced nanoconfinement effects, restrict nutrient diffusion and microbial mobility, resulting in limited degradation despite favorable thermal conditions. This study confirms that intramolecular-level isotopic techniques provide powerful tools for identifying microbial processes in low-permeability shales and contribute to a better understanding of shale oil and gas reservoir evolution.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5497-5508"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859058","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":"Diversity and distribution pattern of high-quality dolostone reservoirs within sequence framework of the Xiaoerbulake carbonate ramp in Bachu area, Tarim Basin, NW China","authors":"Ling Zhou,&nbsp;Guang-Wei Wang,&nbsp;Fang Hao,&nbsp;Ke-Yu Liu,&nbsp;Ping Wang,&nbsp;Rui Xu","doi":"10.1016/j.petsci.2026.06.034","DOIUrl":"10.1016/j.petsci.2026.06.034","url":null,"abstract":"<div><div>The deep to ultra-deep Cambrian strata in the Tarim Basin are widely developed with dolostones, characterized by considerable thickness and an excellent source–reservoir–caprock assemblage, possessing abundant oil and gas resources. This study focuses on the petrology of the carbonate ramp dolostones to clarify the lithofacies associations of the Lower Cambrian Xiaoerbulake Formation. By integrating the sequence stratigraphic framework with multi-scale sedimentary cycle analysis, the types, spatial distribution patterns, and controlling factors of high-quality dolostone reservoirs were systematically investigated. Three types of dolostone were identified: dolograinstone (D1), microbial dolostone (D2), and crystalline dolostone (D3). The pore types of these dolostones are dominated by fabric-selective intergranular pores, intragranular pores, and microbial-related framework pores, exhibiting a notable correlation with lithofacies. The Xiaoerbulake Formation is stratigraphically divided into two transgressive–regressive 3^rd-order sequences and seven 4^th-order subsequences. Five sets of high-quality reservoirs developed vertically, primarily occurring within the 4^th-order regressive cycles. At the meter-scale, a strong vertical cyclic stacking pattern of pores was observed for microbial mound–shoal complexes. Additionally, penecontemporaneous karstification-related dolostones, characterized by excellent quality and considerable thickness, were initially proposed as being controlled by the boundary of 4^th-order to 3^rd-order sequences, affecting approximately 60 m below the 3^rd-order sequence boundary. Four types of high-quality reservoir lithofacies associations were identified, along with a proposed model that elucidates their distribution patterns within inner ramp facies of the Bachu area. In the carbonate ramp, the development of large-scale, high-quality reservoirs is controlled by a combination of factors, including sedimentary microfacies, micropaleogeomorphy, sea-level fluctuations, and early dolomitization. The microbial mound–shoal complex within the inner ramp, the lagoon-margin shoal, and the restricted supratidal flat located at the micropaleogeomorphic heights provide a geological foundation for the formation of high-quality reservoirs. This is primarily attributed to their conducive exposure and dissolution by meteoric water, forming abundant fabric-selective pores at the penecontemporaneous stage.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5271-5291"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859063","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":"Unlocking residual oil after CO2 flooding: The role of tributyl citrate as a solvent-based mobilizing agent","authors":"Zhi-Hao Zhang,&nbsp;Ming-Wei Zhao,&nbsp;Zhong-Zheng Xu,&nbsp;Qin-Fang Ding,&nbsp;Yu-Xin Xie,&nbsp;Ying-Nan Wang,&nbsp;Yi-Zheng Zhang,&nbsp;Huan Zhang,&nbsp;Xu-Guang Song","doi":"10.1016/j.petsci.2026.06.033","DOIUrl":"10.1016/j.petsci.2026.06.033","url":null,"abstract":"<div><div>CO₂ flooding in light-oil reservoirs can induce compositional redistribution and heavy-fraction deposition, leading to pore-structure deterioration, permeability reduction, and limited oil recovery. Although tributyl citrate (TBC) is a promising solvent-type additive for improving oil fluidity and CO₂–oil compatibility, its role in remobilizing residual oil affected by deposition remains insufficiently understood. This study integrates visual interfacial observations and PVT measurements with core-flooding tests and multiscale characterization to elucidate CO₂–oil interaction-driven deposition and the TBC-assisted remobilization process. Results show that CO₂ preferentially extracts light hydrocarbons (C₁–C₉) while enriching intermediate (C₁₀–C₂₉) and heavy components (C₃₀₊), thereby destabilizing crude-oil colloidal structures and promoting irreversible deposition. In homogeneous cores, miscible CO₂ flooding at 20 MPa achieves a high recovery of 77.08% but also induces 15.28% heavy-component deposition and up to 38.23% permeability reduction. In contrast, fractured cores exhibit less deposition but markedly lower sweep efficiency. The addition of TBC mitigates these adverse effects by suppressing viscosity growth, enhancing CO₂ solubility and oil swelling, and improving CO₂–oil compatibility. These synergistic effects strengthen interfacial mass transfer and provide an additional 13%–18% recovery enhancement beyond CO₂ flooding alone. Multiscale observations suggest that TBC reduces the aggregation tendency of deposited heavy components and improves pore-scale flow continuity. These findings provide insight into the remobilization of deposition-affected residual oil after CO₂ flooding and demonstrate the potential of TBC to mitigate formation damage and enhance CO₂-EOR performance in light-oil reservoirs.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5788-5799"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859531","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":"Thermally activated self-adhesive flexible granules for sealing fractures in high-temperature and high-salinity reservoirs: Synthesis, tunability, and performance evaluation","authors":"Yan Xue,&nbsp;Xiao-Feng Zhao,&nbsp;Qiu-Lu Yin,&nbsp;Jing-Yang Gu,&nbsp;Bing Wei,&nbsp;Jun Lu","doi":"10.1016/j.petsci.2026.06.019","DOIUrl":"10.1016/j.petsci.2026.06.019","url":null,"abstract":"<div><div>To overcome the limitations of conventional preformed particle gels in ultra-high-temperature and high-salinity fractured reservoirs, this study developed a novel self-adhesive flexible granule (SAFG). SAFG is synthesized through a two-step process involving the hydrophobic modification of polyvinyl alcohol, followed by chemical cross-linking with terephthalaldehyde, resulting in a stable granular structure. Comprehensive characterization confirmed the successful synthesis, tunable granule size controlled by stirring speed, and exceptional long-term thermal stability in high-salinity brine (2.1 × 10⁵ mg/L, 130 °C for 180 d). Response surface methodology (RSM) optimization determined the optimal synthesis parameters, yielding granules with superior comprehensive performance. SAFG exhibits a unique, tunable self-adhesion mechanism activated by elevated temperature in high-salinity brine, transforming into a cohesive bulk gel. Core-flooding experiments demonstrated effective migration and dynamic plugging in simulated fractures (fracture widths of 0.5 and 1.0 mm), with optimal performance observed at injection rates of 1.5–2.5 mL/min and SAFG concentrations of 5.0 wt%. Critically, after thermally assisted adhesion aging, the adhered SAFG structure achieved a breakthrough pressure of 3080 kPa, approximately 6.3 times the initial stabilization pressure, confirming its robust sealing capability. This work demonstrates SAFG as a robust and tunable conformance control agent for extreme reservoir conditions, offering a viable solution to enhance oil recovery in challenging fractured formations.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5757-5769"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859586","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":"Formation drillability characterization method based on multi-scale convolutional neural representation with stacking ensemble","authors":"Jian-Sheng Liu,&nbsp;Hua-Lin Liao,&nbsp;Zhe Huang,&nbsp;Feng-Tao Qu,&nbsp;Fang Shi,&nbsp;Tian-Yu Wu","doi":"10.1016/j.petsci.2026.04.054","DOIUrl":"10.1016/j.petsci.2026.04.054","url":null,"abstract":"<div><div>Formation drillability characteristics are critical for optimizing drilling performance and designing bottom-hole assemblies. Accurate estimation of drillability parameters, including uniaxial compressive strength (<em>σ</em>_c), drillability grade (<em>K</em>_d), and abrasiveness index (<em>I</em>_a), is fundamental for safe and efficient drilling operations. To overcome the generalization constraints of conventional single models and their inability to capture multi-scale feature responses, this study proposes a novel data-driven framework, termed the multi-scale convolutional neural representation with stacking ensemble (MSCNR-SE), for accurately estimating <em>σ</em>_c, <em>K</em>_d, and <em>I</em>_a. The framework uses a three-stage strategy: (1) a multi-scale convolutional module extracts features across different spatial scales; (2) an ensemble of diverse base regressors (e.g., random forest (RF), gradient boosted decision tree (GBDT)) enhances representational capacity; and (3) extreme gradient boosting (XGBoost) is employed as a meta-learner to integrate the outputs of base regressors, thereby mitigating the risk of model-specific bias. This hierarchical deep-learning ensemble effectively captures complex nonlinear relationships intrinsic to formation properties. Experimental results show that MSCNR-SE accurately models depth-aligned drillability profiles, with strong agreement between predictions and measurements (the coefficient of determination, <em>R</em>² &gt; 0.93 on the blind testing sequence). The MSCNR-SE model achieved superior predictive performance compared with other models across all three parameters. Specifically, MSCNR-SE reduced the mean absolute error (MAE) by approximately 3%–50%, the root mean squared error (RMSE) by 4%–64%, the mean absolute percentage error (MAPE) by 3%–59%, and increased the <em>R</em>² by 1%–32%. These results highlight the superior predictive accuracy and robustness of MSCNR-SE, underscoring its potential as a practical, highly efficient data-driven solution for real-time parameter estimation, monitoring, and control in complex drilling.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5524-5542"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859056","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":"Stress and frequency dependence of wave reflection and transmission in patchy-saturated porous media","authors":"Jia-Yun Li,&nbsp;Zhao-Yun Zong,&nbsp;Xing-Yao Yin,&nbsp;Zhi-Wei Miao,&nbsp;Shi-Kai Li,&nbsp;Wei Xiao","doi":"10.1016/j.petsci.2026.04.032","DOIUrl":"10.1016/j.petsci.2026.04.032","url":null,"abstract":"<div><div>Partially saturated porous media are frequently encountered in the subsurface. However, the wave reflection and transmission (R/T) characteristics of such media under in situ stress are not fully understood. To address this issue, we derive an exact R/T coefficient equation for an inhomogeneous plane wave incident obliquely at the interface between two dissimilar stressed patchy-saturated porous media, based on our new stress- and frequency-dependent rock physics model. The new model simultaneously incorporates the wave-induced fluid flow (WIFF) mechanisms at macroscopic, mesoscopic, and microscopic scales. Specifically, the nonlinear deformation of cracks is incorporated into the microscopic squirt-flow and mesoscopic patchy-saturated models to modify the solid frame and fluid moduli, while the poro-acoustoelasticity (PAE) theory is incorporated to account for macroscopic global flow and in situ stress effects. The modeling results indicate that P- and S-wave dispersion and attenuation decrease with increasing effective stress, but increase with rising water saturation. The opposite effects can be attributed to the closure of compliant cracks and the increased viscosity of the pore fluid, respectively. By comparing our model with published models and experimental data, we confirm its validity and reliability. Subsequently, the variations of the R/T coefficients with incident angle, effective stress, frequency, inhomogeneity angle, and water saturation are further analyzed. The findings demonstrate that the R/T coefficients exhibit significant dependence on effective stress and water saturation across the entire range of incidence angles, whereas the effect of inhomogeneity angle occurs primarily near the critical angle. With increasing effective stress, the critical angle grows, and at small incidence angles, fast P-wave reflection decreases while transmission increases, these effects diminishing at higher stress. The energy conservation at the interface further illustrates the validity of our R/T coefficient equation.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5474-5496"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859057","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":"Differential inorganic–organic porosity evolution in the Wufeng–Longmaxi Shale, southern Sichuan Basin, China","authors":"Guang-Shun Xiao,&nbsp;Neng-Wu Zhou,&nbsp;Qin-Hong Hu,&nbsp;Sheng-Xian Zhao,&nbsp;Bo Li,&nbsp;Bo Song,&nbsp;Xin-Yu Jiang,&nbsp;Zi-Yi Wang,&nbsp;Shuang-Fang Lu","doi":"10.1016/j.petsci.2026.06.058","DOIUrl":"10.1016/j.petsci.2026.06.058","url":null,"abstract":"<div><div>Quantifying how inorganic and organic porosity co-evolve in shale during burial and later uplift is a key scientific issue for constraining shale-gas occurrence, storage capacity, and enrichment mechanisms. Despite extensive studies, major uncertainties remain regarding (1) quantitatively separating compaction-driven inorganic pore loss from maturity-controlled organic pore generation and (2) restoring porosity under in situ effective stress through geological time. Here we develop an integrated quantitative framework for the Wufeng–Longmaxi shale gas reservoirs in the southern Sichuan Basin, China. Core samples from production wells, in which porosity is primarily controlled by total organic carbon (TOC), were selected to minimize the interference of mineral composition and compaction heterogeneity. Using a TOC–porosity-intercept approach, inorganic porosity was calculated from the intercept of the porosity–TOC linear relationship and its burial evolution was reconstructed with a compaction model constrained by well burial histories. Organic porosity was quantified by extracting organic-matter surface porosity from field-emission scanning electron microscopy (FE-SEM) images using ImageJ, and a maturity-dependent organic porosity evolution model was established from samples spanning different thermal maturities. Total porosity evolution in the Luzhou, Changning, Weiyuan, and Western Chongqing blocks was reconstructed by integrating the inorganic- and organic-porosity models, and in situ porosity was further restored using pressure-confined porosity experiments. Results indicate that inorganic porosity decreases sharply with burial depth, from an initial value of ∼50% to 2.47%–4.05% at maximum burial, mainly controlled by mechanical compaction. Organic porosity increases after entering the hydrocarbon generation window and peaks at 1.54%–1.79%, reflecting pore generation during organic matter transformation. Consequently, total porosity declines to 4.16%–5.76% with increasing burial depth, demonstrating that compaction outweighs hydrocarbon-generation-related porosity creation. Restored in situ porosity fluctuates slightly during subsidence but exhibits an overall decreasing trend, reaching 2.34%–3.78% at maximum burial, and then shows only a modest rebound during late-stage tectonic uplift to 2.49%–4.17% at present. Uncertainties are quantified via bootstrap/Monte Carlo propagation and reported as 95% prediction intervals. These results highlight the coupled controls of compaction, hydrocarbon generation, and tectonic unloading on shale pore evolution and provide a quantitative basis for predicting the distribution of high-quality shale reservoirs.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5319-5333"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859082","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":"Physics-informed machine learning for sustained casing pressure diagnostics and root cause analysis for well integrity","authors":"Ahmed Alsubaih,&nbsp;Evan Wetmore,&nbsp;Watheq J. Al-Mudhafar,&nbsp;Kamy Sepehrnoori,&nbsp;Alberto L. Manriquez,&nbsp;Mojdeh Delshad","doi":"10.1016/j.petsci.2026.03.049","DOIUrl":"10.1016/j.petsci.2026.03.049","url":null,"abstract":"<div><div>Sustained casing pressure (SCP) is a primary indicator of well integrity degradation, arising from compromised barriers such as cement, casing, tubing/packer components, or wellhead seals. Traditional SCP diagnostics—based on manual interpretation of annular pressure trends, bleed-off tests, and operational records—are often time-consuming, analyst-dependent, and difficult to scale across large well populations. This study presents a physics-informed machine learning (PIML) framework that integrates engineering-based physical principles, including fluid compressibility, thermal expansion, and leak-path mechanics, into a machine learning workflow for automated and scalable SCP classification. The framework is applied to field monitoring data from 26 wells, with detailed multi-annulus case studies for wells A5, A8, and A12. The method classifies cycle-level SCP behavior into six diagnostic types—no pressure, thermal pressure, trapped pressure, recharge pressure, constant pressure, and high-rate recharge—and maintains physically plausible and interpretable outputs under noisy or complex pressure signatures. Operationally, the framework supports real-time, SCADA-integrated surveillance to enable early detection of integrity threats, prioritization of higher-risk wells, and proactive intervention planning. The novelty of this work lies in embedding physical constraints directly into the classification logic, producing a robust and interpretable diagnostic capability that advances SCP analysis from a reactive, manual task toward automated well-integrity surveillance applicable to offshore, HPHT, CO₂ sequestration, and hydrogen storage operations.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5636-5647"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859579","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":"Cooperative optimization of well-spacing and fracturing in tight reservoirs","authors":"Wen-Sheng Wu,&nbsp;Bo-Wei Zhang,&nbsp;Shi-Kai Tong,&nbsp;Xiu-Kun Wang","doi":"10.1016/j.petsci.2026.04.017","DOIUrl":"10.1016/j.petsci.2026.04.017","url":null,"abstract":"<div><div>The recovery factor of tight oil reservoirs is comprehensively controlled by formation geological properties, well spacing, and fracturing intensity. Insufficient formation drainage or fracture-hits between wells can lead to low recovery factors. The integrated optimization of well spacing and fracturing intensity is a crucial approach to improving recovery factors in tight oil reservoirs. This work analyzes the production characteristics of tight oil wells and identifies the main factors affecting productivity applying machine learning methods. An integrated workflow combining fracturing modeling, reservoir simulation, and production optimization is developed and applied to a real field reservoir block to explore the optimal matching relationship between well spacing and fracturing intensity. The research results indicate that the fractured horizontal well productivity is highly correlated with the stable water cut and is influenced by reservoir geological parameters and fracturing intensity. The material balance method can quickly reconstruct water saturation field after fracturing, which improves history matching. Moreover, simple machine learning method, response surface experiments, is used along with our integrated workflow to generate well spacing and fracturing intensity matching charts under different reservoir permeability conditions and provided the corresponding recovery factor prediction correlations. This study offers a fast and efficient method for developing tight oil reservoir with optimized well spacing and fracturing intensity.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5693-5705"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859582","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":"Machine learning–driven prediction of profile control effect in oil reservoir based on relative production difference","authors":"Kun Xie,&nbsp;Zhan-Qi Wu,&nbsp;Xuan-Shuo Tian,&nbsp;Wei-Jia Cao,&nbsp;Xin Song,&nbsp;Er-Long Yang,&nbsp;Kun Yan,&nbsp;Xiang-Guo Lu","doi":"10.1016/j.petsci.2026.04.021","DOIUrl":"10.1016/j.petsci.2026.04.021","url":null,"abstract":"<div><div>Traditional methods for predicting profile control effect have limited accuracy, with deviations of varying degrees between predicted and actual oil production after operation. In order to solve this problem, this study introduces a prediction method of profile control effect based on relative production difference (PRD). Initially, Long Short-Term Memory (LSTM) network, Gated Recurrent Unit (GRU) network, and Transformer neural network were used to predict the benchmark production of water flooding. Subsequently, the study identified factors influencing profile control effect as input features and the relative difference between benchmark production of water flooding and post-treatment actual production as the target output. Then three machine learning models were constructed for prediction of profile control effect, namely Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), and Random Forest (RF). Hyperparameters of all models were optimized using the Optuna framework during training process. Finally, the study applied the optimal profile control effect prediction model to predict the profile control effect in S reservoir. The results show that LSTM performs the best in predicting the benchmark production of water flooding, with <em>R</em>² values of 0.9874 and 0.9508 for the training and test sets, respectively. RF performs the best in predicting the profile control effect, with <em>R</em>² values of 0.9988 and 0.9995 for the test and cross validation sets, respectively. In oilfield applications, the test set <em>R</em>² of RF exceeds 0.9, which proves that the predicted production after profile control is highly consistent with the actual production, and the model has strong generalization ability.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"23 9","pages":"Pages 5706-5722"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859584","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}