Iranian Journal of Oil and Gas Science and Technology

Abstract Experimental analysis and numerical simulation of CO₂ injection in oil and gas reservoirs are essential for CO₂ sequestration and enhanced oil recovery applications. However, experimental approaches are often expensive and time-consuming, while numerical methods require input parameters such as two-phase relative permeabilities. In this study, two-phase flow simulations are conducted using a numerical solver based on the phase-field approach under steady-state conditions to determine relative permeability curves. Specifically, flow simulations are performed in realistic porous media, and nonconstant fluid viscosity and density are incorporated, addressing limitations of previous studies. Therefore, the main contribution of this work is the determination of CO₂–oil relative permeabilities in realistic porous media extracted from micro-CT images under conditions of variable fluid viscosity and density. In addition, the influences of wetting properties and hysteresis on relative permeability behavior are investigated.
The numerical results indicate that increased wettability toward the injected fluid enhances displacement efficiency by shifting the intersection point of the relative permeability curves to the right, leading to higher oil recovery, particularly at larger contact angles corresponding to non-wetting conditions and at lower surface tensions. Furthermore, the separation between imbibition and drainage relative permeability curves, representing hysteresis effects, becomes more pronounced as wettability shifts toward larger contact angles. These findings demonstrate the capability of steady-state pore-scale simulations based on the phase-field approach to reliably determine CO₂–oil relative permeability curves when realistic porous media and nonconstant fluid properties are considered.

Abstract Due to their geological complexity, the Persian Gulf gas fields require more in-depth investigation. The primary reservoir rocks are carbonate sediments, notably the Upper Permian Dalan and Lower Triassic Kangan Formations. The Dalan Formation consists predominantly of oolitic limestone successions, which conformably overlie the Faraghan Sandstone Formation and underlie, unconformably, the carbonate sediments of the Kangan Formation. The Kangan Formation, in turn, underlies the shaly Agar Formation. The boundary between these studied formations corresponds to the Permian–Triassic (P/T) global boundary. In this study, two critical reservoir parameters—porosity and permeability—of the Dalan and Kangan Formations were evaluated using the Nuclear Magnetic Resonance (NMR) method. A thorough understanding of reservoir porosity and permeability is essential for optimal field management and prolonging the productive lifespan of the gas field.
Based on petrophysical properties, the studied succession has been subdivided, from top to bottom, into four units, designated K1 through K4. The results revealed significant differences in the reservoir characteristics of each unit. Units K1 and K2 belong to the Kangan Formation, while units K3 and K4 correspond to the Dalan Formation. Unit K2 exhibits the lowest hydrocarbon potential in terms of reservoir quality, whereas unit K4 of the Dalan Formation shows the highest porosity and permeability values. The permeability of the studied sequence was estimated using three approaches derived from NMR data: the Schlumberger-Doll Research (SDR) model, the T2 (so-called) model, and the Free Flow and Swanson models, based on the calculated porosity. Comparison of the results indicated that the SDR model provides the most reliable permeability estimates.
For sedimentary modeling, the accurately determined porosity and permeability parameters were integrated with geological interpretations. The analysis of the upper and lower deposits across the P/T boundary indicates that erosional events significantly affected reservoir quality. The findings of this study can aid in predicting reservoir performance and reducing drilling risks within the studied basin.

Abstract Residual oil saturation (Sor) estimation is a critical component of reservoir development and enhanced oil recovery (EOR) projects. Traditionally, Sor is estimated using the Archie method, in which accuracy strongly depends on petrophysical exponents, including the tortuosity factor (a), cementation factor (m), and saturation exponent (n). However, the conventional method assumes a homogeneous rock formation, rendering it ineffective and unreliable in shaly sand reservoirs. Additionally, field-based determination of petrophysical exponents and Sor is often difficult and time-consuming. This study addresses these limitations by developing a Python-based application that integrates a three-dimensional (3-D) regression technique with the Newton–Raphson algorithm. The application was tested using well logs from eight wells and statistically validated against core Sor data from the FAS Field, Offshore Niger Delta. Results indicated that the tortuosity factor (a) ranged from 0.28 (FAS-06) to 2.73 (FAS-04), the cementation factor (m) varied from 0.43 (FAS-06) to 4.11 (FAS-04), and the saturation exponent (n) ranged from 0.71 (FAS-04) to 8.59 (FAS-05). Correspondingly, Sor ranged from 0.11 (FAS-05) to 0.99 (FAS-06). The percentage deviation of the computed Sor relative to the core data ranged from 5% (FAS-01) to 27% (FAS-02) for the 3-D regression method and from 3% (FAS-03) to 52% (FAS-02) for the Newton–Raphson technique. The results indicate that the 3-D regression method is more efficient and reliable for computing petrophysical exponents and Sor in the study area.

Abstract Lost circulation of drilling mud is one of the most significant challenges during drilling operations, leading to time loss and increased costs. It is influenced by factors such as the lithology and composition of the formation, the pressure difference between the drilling mud hydrostatics and the formation, the type of drilling mud, and hydraulic conditions. In the Asmari Formation of the Maroun Oil Field in Iran, lost circulation commonly occurs due to natural fractures, cavernous structures, and drilling-related factors. Several methods exist to mitigate this problem, including the use of appropriate loss circulation materials (LCMs) and reducing the drilling mud density. In this study, oyster shell and nanographene are proposed as LCM additives for optimized oil-based drilling mud in the Maroun Oil Field, and the necessary experiments were conducted. This paper investigates the simultaneous use of oyster shell as a normal particle-sized material and nanographene as a nanoparticle-sized material to control mud loss. The results indicate that nanographene is unable to seal fractures larger than 0.04 inches, making its use for wider fractures uneconomical. Oyster shell is ineffective as an LCM for fractures wider than 0.08 inches but performs well in sealing fractures smaller than 0.08 inches.

Abstract The purpose of this study is to apply finite element analysis to identify the stress levels of a standard chemical tower subjected to explosive loads. The geometric finite element model is developed based on the design drawings of the tower, and a mathematical model of the explosion load is defined in accordance with relevant specifications, including ASCE 41088 and recent literature. By accounting for geometric large deformation and material nonlinearity of the tower, the time histories of stress and deformation under explosion loading are obtained, and the analysis results are subsequently validated. Two explosion scenarios are examined: a moderate case with a peak side-on overpressure (P_so) of 14.6 kPa and a severe near-field case with P_so of 30 kPa. Comparative results indicate that increasing explosion intensity leads to significant increases in both deformation and stress, with pronounced stress concentrations consistently observed at the bottom of the tower. Numerical damping is introduced to investigate its mitigation effects, and the results confirm that damping effectively reduces peak structural responses under high-intensity loading. A buckling analysis shows that the first buckling mode initiates at the skirt, identifying this region as particularly vulnerable to instability. Furthermore, bolt deformation and stress remain within safe limits throughout the explosion events. The functional relationship between peak reflected pressure, action time, and spatial position under explosion loading is also established. This study provides an important theoretical basis and technical support for the anti-explosion design and safety assessment of the tower. The results contribute to improving the safety and reliability of chemical plants equipped with such towers, particularly under explosion accident scenarios.

Abstract Machine learning is emerging as a transformative force in reservoir engineering, addressing long-standing challenges in hydrocarbon exploration and production with unprecedented speed and accuracy. This review examines the core machine learning paradigms—supervised learning, unsupervised learning, and reinforcement learning—applied across eight critical domains: reservoir simulation, pore pressure prediction, history matching, reservoir characterization, acidizing operations, hydraulic fracturing, waterflooding, and CO₂-enhanced oil recovery. By leveraging algorithms such as artificial neural networks, support vector machines, particle swarm optimization, and long short-term memory networks, machine learning achieves up to three orders of magnitude improvement in computational efficiency and predictive accuracy exceeding 90%, with the potential to approach 100% under optimal conditions. These performance levels substantially surpass those of traditional methods, which typically plateau at approximately 75% accuracy. Machine learning models integrate heterogeneous, multi-scale data to optimize real-time operations, enhance decision-making, and significantly reduce operational costs. Furthermore, machine learning contributes to sustainability through energy-efficient computational frameworks and enhanced carbon sequestration capabilities. Although challenges such as data scarcity and limited model interpretability persist, emerging approaches—including transfer learning and physics-informed models—are creating new opportunities. Overall, this review underscores the capacity of machine learning to redefine reservoir engineering toward a more intelligent and sustainable energy future.