Petroleum Research最新文献

{"title":"Comprehensive judgment method of low production causes in tight gas reservoirs——Taking the LX gas reservoir in Ordos Basin as an example","authors":"Ren-Shi Nie,&nbsp;Peidong Qiu,&nbsp;Jingcheng Liu,&nbsp;Bin Liu,&nbsp;Zhangxin Chen,&nbsp;Cong Lu,&nbsp;Fan-Hui Zeng","doi":"10.1016/j.ptlrs.2025.11.002","DOIUrl":"10.1016/j.ptlrs.2025.11.002","url":null,"abstract":"<div><div>The low production of tight gas reservoirs arises from complex and multifaceted causes, and elucidating these factors is crucial for guiding the formulation of effective stimulation strategies. Conventional analytical methods predominantly emphasize the role of individual factors, thereby lacking necessary systematic and integrative perspective to comprehensively reveal underlying mechanisms of poor well performance. To overcome these limitations, a comprehensive diagnostic approach is proposed to identify controlling factors of low production in multilayer tight gas reservoirs. Taking three typical cluster well groups of LX tight gas reservoirs in the Ordos Basin as examples, the causes for the low production of the gas wells were analyzed from different aspects by using geological, engineering, and developmental data after excluding the special reasons such as defects in drilling and completion process, reservoir water lock, and water flooding. The results show that the main causes for the low production of gas wells in LX reservoirs include fewer exploited gas layers, poor physical properties, poor fracturing and fracture making effects, and insufficient formation energy. In addition, corresponding treatment measures such as reperforation and refracturing were proposed. Finally, effects of the measures in the two low-yield wells where the reperforation was implemented were analyzed: the production and Tubing-head pressure increased significantly, and the low-yield wells were transformed into non-low-yield wells with good results. The results of the measures demonstrate the feasibility of the new method.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 2","pages":"Pages 529-544"},"PeriodicalIF":3.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148501867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated approach and insights into shale gas potential: Geological, petrophysical, geochemical, and geomechanical evaluation of the Khatatba Formation (Western Desert, Egypt)","authors":"Mohamed I. Abdel-Fattah,&nbsp;Hamdan A. Hamdan,&nbsp;Adnan Q. Mahdi,&nbsp;Zakaria M. Abd-Allah,&nbsp;Nezar A. Hammouri,&nbsp;Sara M. Abuzied","doi":"10.1016/j.ptlrs.2025.11.007","DOIUrl":"10.1016/j.ptlrs.2025.11.007","url":null,"abstract":"<div><div>This study provides a comprehensive assessment of the shale gas potential of the Middle Jurassic Khatatba Formation in the Obaiyed Field, Western Desert, Egypt, by integrating geological, petrophysical, geochemical, and geomechanical datasets into a unified workflow. The multidisciplinary approach allows for the identification of shale gas “sweet spots” and addresses critical operational risks associated with drilling and completion. Petrophysical evaluation indicates heterogeneity, with total porosity ranging from 6 to 12%, effective porosity between 4 and 8%, shale volume often exceeding 40% but locally decreasing below 30%, and permeability within 0.01–0.1 mD. Gas storage capacity is enhanced at depths greater than 3800 m, where free gas contents are significant and adsorbed gas reaches 1.5–2.0 cm³/g in Total Organic Carbon (TOC)-rich intervals. Geochemical analysis confirms that the Upper Safa Member is thermally mature within the gas window, with TOC averaging ∼4 wt% and mixed Type II–III kerogens, while traces of CO₂ (0.8–2.0 mol%) and H₂S (25–80 ppm) raise concerns about casing corrosion and necessitate careful material selection. Geomechanical results reveal brittle intervals with high Young’s modulus and low Poisson’s ratio, fracture gradients ranging from 25 to 30 MPa/km, and maximum horizontal stress (σHmax) typically 1.2–1.5 times the minimum horizontal stress (σhmin), defining a narrow safe mud weight window. Seismic inversion delineates TOC-rich, low-impedance intervals as optimal drilling targets, while fault-bounded compartments highlight both opportunities for hydrocarbon trapping and risks of reservoir compartmentalization. The integration of reservoir quality (RQ) and completion quality (CQ) with stress and pressure profiles enables optimized well orientation and trajectory planning, particularly recommending horizontal wells perpendicular to σHmax in the lower Upper Safa Member. This integrated evaluation confirms the Upper Safa Member as the most promising shale gas target within the Khatatba Formation and establishes a transferable workflow for unconventional reservoir development worldwide.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 2","pages":"Pages 415-430"},"PeriodicalIF":3.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148501870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Asphaltene deposition during miscible and immiscible gas injection in tight heavy oil reservoir: Implications for formation damage","authors":"N. Shakibasefat,&nbsp;M.R. Malayeri,&nbsp;M. Riazi","doi":"10.1016/j.ptlrs.2025.11.001","DOIUrl":"10.1016/j.ptlrs.2025.11.001","url":null,"abstract":"<div><div>Gas injection is frequently used to displace the trapped oil. However, this may lead to deposition of asphaltene, which can profoundly block the pores and throats of the porous media, resulting in reduced porosity and permeability. The rock lithology, composition of the injected fluids, pressure, and temperature are among the parameters that may influence asphaltene deposition. This study investigated the asphaltene precipitation/deposition while injecting rich hydrocarbon gas into a heavy oil reservoir. The experiments were conducted under two conditions: miscible and immiscible gas injections. Two rich hydrocarbon gases with different compositions were used, and experiments were conducted under reservoir conditions. Experiments were performed in carbonate cores previously saturated with heavy live oil. The extent of formation damage was assessed by measuring asphaltene deposition during the flooding of miscible and immiscible rich hydrocarbon gases into the core plugs. The outlet oil samples from the core flooding were analyzed using molecular weight measurements, UV spectroscopy, and the IP143 method. During the miscible flooding process, where rich hydrocarbon gas was injected into live oil at 210 °F, the molecular weight of the outlet oil samples decreased by 53%. The porosity and permeability of the carbonate core in miscible flooding at 210 °F, after washing with cyclohexane, showed a reduction of 13.6% in porosity and 42.8% in permeability. In the course of immiscible flooding with rich hydrocarbon gas, the outlet oil at 104 °F decreased by 48.3% in molecular weight, 35.5% in absorption coefficient, and 30.7% in asphaltene content. The porosity and permeability of the carbonate core in immiscible flooding decreased by 11.8% and 43.25%, respectively. The results showed that both miscible and immiscible injection processes reduce the porosity and permeability of carbonate cores. The damage was particularly severe during the miscible flooding.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 2","pages":"Pages 516-528"},"PeriodicalIF":3.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148501902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Experimental investigation of pressure, temperature, and salinity effects on the interfacial tension of H2/brine/cushion gases: Implications for H2 geo-storage","authors":"Fatemeh Daraii,&nbsp;Payam Moradi,&nbsp;Seyed Ali Bahreini,&nbsp;Sina Anvari","doi":"10.1016/j.ptlrs.2025.11.004","DOIUrl":"10.1016/j.ptlrs.2025.11.004","url":null,"abstract":"<div><div>Underground hydrogen storage (UHS) in saline aquifers is a promising solution for large-scale, long-duration energy storage in a low-carbon energy system. Cushion gases, such as N₂, CO₂, and CH₄, are commonly used to regulate reservoir pressure and enhance storage efficiency. However, despite recent advancements in understanding the interfacial tension (IFT) behavior of H₂–gas mixtures in contact with brine under reservoir conditions, key gaps persist in systematic data for H₂–cushion gas mixtures (e.g., with N₂, CO₂, and CH₄) across specific thermophysical ranges, despite its influence on capillary sealing, gas mobility, and recovery. This study presents systematic experimental measurements of IFT and density for H₂–N₂, H₂–CO₂, and H₂–CH₄ gas mixtures (70:30 mol%) with both distilled water and brine (30,000 ppm NaCl) at different temperatures (25, 50, and 70 °C) and pressures (1, 5, and 10 MPa). The IFT was measured using a high-pressure, high-temperature drop tensiometer, and density was obtained via an Anton Paar DMA 4500 densitometer. Results show a consistent decrease in IFT with rising temperature and pressure across all systems. Notably, the H₂–CH₄ mixture exhibited an IFT of 69.4 mN/m in distilled water at 25 °C and 1 MPa, decreasing to 57.3 mN/m at 70 °C; in contrast, the same mixture in 30,000 ppm brine showed a larger reduction from 61.3 mN/m to 47.8 mN/m under identical conditions, highlighting the amplifying effect of salinity on IFT decline. Among the gas mixtures, CH₄ yielded the approximately lowest overall IFT values, favoring injectivity, whereas N₂ maintained the highest IFT, supporting enhanced capillary retention. These findings address a critical data gap for realistic cushion gas performance and provide foundational insight for modeling UHS in saline aquifers.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 2","pages":"Pages 545-555"},"PeriodicalIF":3.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148501868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of neutron-X-ray dual-modal imaging in oil and gas well cementing under ultra-high temperature conditions","authors":"Yabing Wen ,&nbsp;Xinxiang Yang","doi":"10.1016/j.ptlrs.2025.11.011","DOIUrl":"10.1016/j.ptlrs.2025.11.011","url":null,"abstract":"<div><div>Understanding the failure mechanisms affecting cement sheath integrity under ultra-high temperature (UHT) conditions is essential for ensuring wellbore integrity and supporting the safe and sustainable extraction of deep oil and gas resources. However, difficulties in characterizing moisture migration have limited understanding of the physicochemical mechanisms and cementing processes that drive failure of cementing materials in such environments. To address these challenges and enhance understanding of cement sheath failure mechanisms, this study systematically evaluates the adaptability and advantages of Neutron–X-ray Dual-modal Imaging (NXDI) technology. NXDI’s extensive applications in materials science, mineralogy, civil engineering, and energy research provide representative evidence of its potential. Building on this foundation, the study proposes, for the first time, a conceptual framework, technical roadmap, and future prospects for applying NXDI to cement sheath research under UHT conditions. Our findings indicate that intensified moisture migration critically affects the microstructure and physicochemical stability of cement sheaths. NXDI offers unique potential to overcome the limitations of conventional imaging methods by simultaneously characterizing moisture dynamics and microstructural evolution. Nevertheless, its application still faces challenges, including neutron imaging quality, dual-modal image fusion, sample representativeness, and practical implementation. To address these, we outline key research priorities and feasible steps. With continued advances in imaging technologies and experimental methods, NXDI is expected to play an increasingly important role in enhancing the efficiency, safety, and sustainability of deep oil and gas extraction.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 2","pages":"Pages 590-612"},"PeriodicalIF":3.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148268489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computation of pressure depletion and well productivity of multi-fractured wells with variable fracture spacing using fast and accurate solution method","authors":"Yakai Tian ,&nbsp;Ruud Weijermars","doi":"10.1016/j.ptlrs.2025.08.008","DOIUrl":"10.1016/j.ptlrs.2025.08.008","url":null,"abstract":"<div><div>This study presents a fast method that can model the pressure depletion around hydraulically fractured wells accurately. The Gaussian Pressure-Transient (GPT) method has infinite resolution, because no discrete-grid computations are required − the method solves the pressure field using closed-form solutions. After extensive testing on field cases and calibration against other (independent) solution-methods, it appeared that the pressure interference of multiple fractures represented by pressure transients − in the early phases of pressure advance − behaves according to the traditional superposition-method. However, when pressure interference advances, the traditional superposition method gives increasingly unrealistic and inaccurate results. Therefore, the traditional superposition method typically used for steady-state, needs adjustment when applied to transient systems, modeled here with Gaussian pressure-diffusion probability-density functions. The new solutions are valid for the advance of pressure changes both in the early and late stages of the well-life. A second new insight presented here is that the superposition of multiple fractures using Gaussian solutions (together with pressure-gradient computations and application of Darcy's) can accurately forecast the future production rate and cumulative production of the well over the economic well-life, accounting for the impact of fracture-spacing changes. The new method is validated by history-matching of our Gaussian forecasts using real well-data to constrain specific reservoir properties (hydraulic diffusivity) and completion parameters (fracture half-lengths), and then was benchmarked against independent simulations with CMG-IMEX.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 1","pages":"Pages 150-162"},"PeriodicalIF":4.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147556434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pseudo-pressure behavior in multilayered reservoir for single phase flow with permeability-pressure dependent","authors":"Jessica Lailla Ferreira Bittencourt Neto,&nbsp;Sinésio Pesco,&nbsp;Abelardo Borges Barreto Júnior","doi":"10.1016/j.ptlrs.2025.07.002","DOIUrl":"10.1016/j.ptlrs.2025.07.002","url":null,"abstract":"<div><div>Mathematical techniques continue to become more advanced in developing more complex solutions in different fields, especially in oil reservoir simulation, where various physical parameters are still unknown. Studying different simulation cases that improve predictability is required to comprehend the wellbore behavior under production properly. This article proposes a new analysis method that utilizes the first-order asymptotic expansion method to address single-phase oil flow in multilayered reservoirs with permeability-pressure dependent and source terms. The classical techniques, like pseudo-pressure functions, although offering small linearization of the diffusivity equation, are found to be insufficient and thus require more efficient techniques like the Perturbation Method. The proposed model allows detailed analysis of different reservoir properties and pressure-dependent permeability curves, therefore allowing for more effective optimization and understanding of the flow behavior in complex geological conditions.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 1","pages":"Pages 50-62"},"PeriodicalIF":4.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147556506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Study on the expansion characteristics of oil-based drilling fluid in HTHP wellbore","authors":"Mingjie Cai,&nbsp;Dan Mao,&nbsp;Yuyao Li,&nbsp;Tingjun He,&nbsp;Mingmin He,&nbsp;Leichuan Tan","doi":"10.1016/j.ptlrs.2025.08.003","DOIUrl":"10.1016/j.ptlrs.2025.08.003","url":null,"abstract":"<div><div>Accurate prediction of oil-based drilling fluid (OBD) behavior under high-temperature, high-pressure (HTHP) conditions in ultra-deep wells is essential for ensuring well control safety. However, research on the volumetric expansion of OBD in HTHP wellbores remains limited. This study develops a mathematical model to predict the expansion characteristics of OBD in HTHP wellbores, incorporating temperature- and pressure-dependent variations in rheological properties, thermophysical parameters, and density. The model is solved using discretization and the successive over-relaxation (SOR) method. Laboratory experiments were conducted to measure the rheology and density of OBD at different initial densities under varying temperature and pressure conditions. Binary mathematical models describing density and rheology as functions of temperature and pressure were established. The results indicate that the rheological properties of OBD exhibit consistent trends across different initial densities. Apparent viscosity (AV) and plastic viscosity (PV) decrease with increasing temperature and gradually stabilize, while both increase with rising pressure. The density of OBD decreases due to thermal expansion at elevated temperatures and increases due to thermal contraction at lower temperatures. Temperature exerts a greater effect on OBD density than pressure. Model validation using field data from an ultra-deep well confirms the accuracy of the proposed approach. During drilling and tripping operations, changes in mud pit volume occur due to temperature fluctuations; however, these changes are relatively slow compared to the rapid volume changes associated with wellbore influx. It is recommended to suspend the use of drilling fluid cooling devices before pulling out of hole (POOH) operations to minimize mud pit volume fluctuations. These findings provide theoretical guidance for accurate bottom-hole temperature prediction and improved drilling strategy optimization.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 1","pages":"Pages 237-249"},"PeriodicalIF":4.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147556439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Oil flow impact characteristics of internal floating roof tanks during floating roof landing and oil receiving processes","authors":"Yan Li ,&nbsp;Chunqiang Zhu ,&nbsp;Fulin Yang ,&nbsp;Xinpeng Wang ,&nbsp;Jieyao Jia","doi":"10.1016/j.ptlrs.2025.05.006","DOIUrl":"10.1016/j.ptlrs.2025.05.006","url":null,"abstract":"<div><div>To maintain the integrity of internal floating roof tanks, curb environmental pollution and guarantee the safety of tank farms, it is important to determine the oil flow features that impact the floating roof during the oil receiving process, especially when the floating roof falls to the bottom of the tank. In this study, the finite volume, volume of fluid (VOF), and realizable <em>k-ε</em> turbulence models and the pressure implicit with splitting of operators (PISO) algorithm were adopted to establish a three-dimensional physical model of an internal floating roof tank and simulate the oil receiving process. A pressure test platform was then set up to validate the model's viability. The flow field characteristics and influencing properties of the impact of the liquid against the floating roof under different oil inlet velocities, initial liquid levels, medium densities, and tank diameters were analyzed and compared. The results showed that increasing the oil inlet velocity and medium density led to the maximum pressure and the impact force at different moments of the liquid against the floating roof also increasing. An increase in the oil inlet velocity, led to greater turbulence on the opposite the inlet, and higher pressure at the edge of the floating roof in the upper area opposite the inlet. At an oil inlet velocity of 4.0 m/s, the maximum pressure was 135.5 kPa and the maximum impact force peak was 13.967 MN. The initial liquid level increase had little influence on the liquid's impact on the floating roof, and the high pressure zone was at the edge of the floating roof opposite the inlet. When the initial level was 0.5 m, the maximum pressure was 135.81 kPa and the maximum impact force peak was 14.021 MN. As the density of the medium increased, the greater the turbulence on the inlet relative side and the higher pressure at the edge of the floating roof in the upper area of the inlet relative side. When diesel was the medium, the maximum pressure was 154.9 kPa and the maximum impact force peak was 15.999 MN. As the diameter of the tank increased, the maximum pressure of the floating roof impacted by the liquid reduced, but the impact force increased. The slower the turbulence on the inlet relative side, the greater the maximum pressure zone of the floating roof. When the tank volume was 1000 m³, the maximum pressure was 135.5 kPa. When the tank volume was 10,000 m³, the maximum impact force peak was 81.978 MN. Considering the above, a low velocity, a low-density light oil, and an internal floating roof tank with a small diameter are recommended for engineering operations that involve receiving oil. Further increased safety precautions at the edge of the floating roof opposite the inlet are needed. In sum, the findings of the study offer theoretical justifications and a technical reference for optimizing the structural design of floating roofs and the oil receiving and delivering process.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 1","pages":"Pages 297-308"},"PeriodicalIF":4.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147556508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimation of fluid saturation and pressure distribution throughout a reservoir using machine learning techniques","authors":"Arifur Rahman ,&nbsp;George Daoud ,&nbsp;Ezeddin Shirif ,&nbsp;Mohamed El-Darieby ,&nbsp;Mohamed El-Hendawi","doi":"10.1016/j.ptlrs.2025.08.007","DOIUrl":"10.1016/j.ptlrs.2025.08.007","url":null,"abstract":"<div><div>Water saturation is one of the most critical yet often underappreciated petrophysical parameters in reservoir characterization. A wide range of petrophysical and reservoir engineering computations that lead to crucial field development decisions, including reserve estimation, waterflooding efficiency calculation, and capillary pressure deduction, rely on its accurate determination. This study demonstrates how machine learning techniques can forecast reservoirs’ fluid saturation and pressure distribution. This study describes a deep learning–based proxy modeling technique for accurately predicting reservoir pressure distribution and fluid (oil, water, and gas) saturation during water flooding in single-layer heterogeneous reservoirs. This study used recurrent neural networks (RNNs) and convolutional neural networks (CNNs) to build a proxy model. This work indicates that compared to simulation outcomes, computer-based machine learning algorithms can accurately predict fluid (oil, water, and gas) saturation and pressure distribution. The stated accuracy was evaluated numerically and graphically, and error analysis between various machine learning approaches and simulated results was utilized.</div></div>","PeriodicalId":19756,"journal":{"name":"Petroleum Research","volume":"11 1","pages":"Pages 217-236"},"PeriodicalIF":4.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147556509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}