https://pr.ripi.ir/ Journal of Petroleum Research en daily 1 Tue, 21 Apr 2026 00:00:00 +0330 Tue, 21 Apr 2026 00:00:00 +0330 https://pr.ripi.ir/article_1557.html Achieving methane production from a gas hydrate-containing system under safe and economic conditions requires a fundamental understanding of the physical behavior involved in the dynamic process of methane hydrate formation and separation in sedimentary environments. In this study, by designing a laboratory device, the kinetics of methane hydrate formation in a porous silica environment with an excess water approach in offshore fields were simulated. Also, using volume balance calculations, the methane conversion fraction during different stages of hydrate formation and the saturation of phases in hydrate-containing sediments were obtained. The results of the temperature and pressure evolution diagrams during the methane hydrate formation process showed that the hydrate formation conditions for laboratory samples are in great agreement with the prevailing thermodynamic conditions. In proportion to the selected silica sand grain size, the average saturation of the hydrate phase was 32.5%, the saturation of the water phase was 63.9%, and the saturation of the gas phase was 3.6% for laboratory samples. Furthermore, the average methane to hydrate conversion fraction for test samples was 88.2%. Moreover, the difference in the thermal responses of the system at some positions of the thermocouples was attributed to the non-uniform distribution and formation of methane hydrate in the porous medium. Also, the studies showed that the amount of access of gas and water phases and the type of distribution of the hydrate phase in the sedimentary medium are important factors controlling the methane conversion fraction in the experimental samples. The simulation of the temperature and pressure conditions of the gas hydrate formed in this study also had a good agreement with gas hydrate reservoirs in offshore fields. https://pr.ripi.ir/article_1555.html Geomechanical parameters play a key role in the oil and gas industry by assisting in the evaluation of reservoir rock properties, predicting formation behavior during drilling, and optimizing hydrocarbon production. In this study, geomechanical parameters were used to identify the most favorable hydrocarbon zones in an oilfield and to determine reservoir lithology. In the first stage, hydrocarbon-prone areas were identified using an integrated attribute defined as the product of geomechanical parameters and acoustic impedance, obtained from pre-stack seismic inversion. A decrease in the value of this composite attribute indicates a higher probability of hydrocarbon presence in the area. Subsequently, isochore maps were generated for the identified areas, and zones with the greatest thickness were selected as optimal targets for exploration, extraction, and production. After determining the precise drilling locations, the reservoir lithologies—including shale, limestone, gas-bearing limestone, and oil-bearing sandstone, which had been previously identified from drilling cuttings—were analyzed and differentiated using various geomechanical scatter plots. These plots included Young’s modulus versus Poisson’s ratio, Zₚ versus Vₚ/Vₛ, Mu-Rho versus Lambda-Rho, and Poisson’s ratio versus Lambda-Rho. The purpose of these plots was to accurately examine lithological variations based on seismic data and to enable reliable identification and classification of lithologies without the need for drilling. This approach provides a framework for future studies to enhance lithological identification by employing similar plots and consistent geomechanical trends. To assess the accuracy of the method, the compressional impedance and Young's modulus derived from seismic data were compared with corresponding well log values. A 90% match in acoustic impedance and over 80% agreement in Young's modulus confirm the method's accuracy and reliability in characterizing the geomechanical properties of the reservoir. https://pr.ripi.ir/article_1577.html In thermal methods of oil extraction from oil shales, kerogen is converted into liquid and gaseous hydrocarbons by heating to high temperatures (usually 300 to 500 degrees Celsius). Moreover, oil extraction is accompanied by a solid residue (ash) containing minerals, heavy metals, and inorganic compounds. Ash, as a by-product, has various effects on the extraction process and the environment and plays an important role in the process of oil extraction from oil shales. Furthermore, water, as a strong absorber of microwaves, can act as a «heat carrier» in the extraction process and improve heat transfer to kerogen. In this study, the effect of ash and moisture on the characteristics of oil extracted from oil shale in heating using microwaves and the conventional method was investigated. In the microwave irradiation process, adding ash in amounts of 25 and 50 weight percent to the sample increased oil recovery. In addition, in the conventional heating process, adding ash was accompanied by an increase and decrease in oil recovery. Moreover, increasing the percentage of carbon and saturated fraction, decreasing the chain length of aliphatic compounds along with decreasing sulfur, asphaltene fraction and decreasing aromatic compounds and its entanglement index were considered as a function of improving the produced oil. The results of the percentage of carbon and sulfur components, saturated and asphaltene fractions and FT-IR, C&H-NMR spectra confirm that in microwave heating with increasing ash weight, although the amount of extracted oil shows a slight increase, the quality of the produced oil decreases. This indicates an optimal amount of ash in the extraction process that depends on its components and amount. For comparison, the experiments were repeated with conventional heating, and the results indicate that ash does not play a constructive role in improving the extracted oil in conventional heating. By adding 2.9% by weight of water to the sample, the presence of moisture in the microwave heating process was evaluated, which caused a decrease in the extracted oil. Qualitatively, the percentage of carbon and saturated fraction had a decreasing trend, sulfur and asphaltene fraction did not change significantly. The results of FT-IR and C&H_NMR spectra also indicated no improvement in the quality of the extracted oil. In conventional heating, although humidity increased the amount of oil produced, no improvement in quality was observed. Therefore, in general, humidity was not beneficial in the oil extraction process. https://pr.ripi.ir/article_1558.html Reconstructing high-resolution images from poor-quality images is one of the basic challenges in image processing. In modeling porous media using images obtained from computed microtomography (Micro-CT), high-quality images of these media are usually unavailable for different reasons, including the cost and computational complexity of high-quality imaging. With the advancement in the development of multiscale networks, it is possible to use more image details in these networks, and there is a further need for high-quality images. Today, adversarial generative networks are used as a practical tool in increasing image quality. These networks are trained using pairs of high-quality and low-quality images and then, they produce high-quality images by taking low-quality images as input. The final images used in modeling the porous media are binary images, but using binary images as input in the image quality enhancement process may result in detail loss. For this purpose, this study investigates the effect of using grayscale or binary images as input in the model training process. Grayscale images of a Brea sandstone sample are taken as the main input and binary images are generated using grayscale images and the realization of the real porosity of the rock sample. In this research, a new model called RealESRGAN is used to enhance the quality of the images. https://pr.ripi.ir/article_1556.html various methods have been proposed for reservoir rock typing; these approaches are often associated with limitations and simplifying assumptions that hinder accurate definition of flow units and dynamic classification of reservoir rocks. Studies indicate that the selection of an appropriate rock typing method depends on the intrinsic properties of the reservoir rock as well as the quantity and quality of the available data. Rock typing is a key strategy for mitigating these challenges by grouping reservoir rocks with similar petrophysical and flow properties. In this study, multiple rock typing methodologies—including the Flow Zone Indicator (FZI), Modified Flow Zone Indicator (FZIM), the Winland method, and permeability-based classifications—were applied to a carbonate reservoir in a southwestern Iranian oil field. To evaluate the effectiveness of these methods, statistical techniques such as the Elbow method and correlation analysis were employed. Among the evaluated models, the FZIM, the permeability-based classification, and the R-model exhibited superior performance based on relative permeability curve analysis. Consequently, these models were selected for use in dynamic reservoir simulation. In addition, two novel rock typing models were developed using relative permeability data as the basis for classification. Simulation results demonstrated that the proposed models achieved an acceptable history match for water production trends. However, rock type classification showed a limited impact on the history matching of gas-oil ratio and reservoir pressure decline. https://pr.ripi.ir/article_1559.html In this study, the treatment of oily wastewater was investigated using a combined coagulation and polyethersulfone (PES) ultrafiltration membrane process. The PES membrane was fabricated via the phase inversion method, and its morphology was examined using scanning electron microscopy (SEM). In the coagulation–ultrafiltration process, ferric chloride (FeCl₃) and polyaluminum chloride (PAC) were applied as coagulants at different concentrations and at pH levels of 3, 5, 7, and 10. These conditions were tested both individually and in combination, in the presence and absence of ultraviolet (UV) irradiation, over a period of 30 minutes, to determine the optimal conditions for chemical oxygen demand (COD) removal. The results indicated that in the coagulation–ultrafiltration process, increasing the coagulant concentration and adjusting the pH significantly improved permeate flux and reduced COD levels. UV irradiation was found to enhance the process efficiency by partially decomposing organic compounds and influencing floc formation, thereby improving both permeate flux and COD removal performance. Overall, the integrated coagulation–PES membrane process, whether applied with or without UV irradiation, led to increased flux rates and enhanced COD removal efficiency. The COD removal efficiencies achieved were 86% for the coagulation process alone and 96% for the combined coagulation–ultrafiltration process. https://pr.ripi.ir/article_1563.html Fractured reservoir rocks are among the most critical targets in petroleum engineering studies, as their complex networks of fractures and cracks play a decisive role in the porosity and permeability of these formations. CT-scan imaging is a key tool for analyzing the internal structure of such rocks; however, reduced image quality often makes accurate reconstruction challenging. In this study, two deep-learning architectures a Convolutional Autoencoder and a Vision Transformer were designed and compared with the objective of enhancing and reconstructing CT-scan images of fractured reservoir rocks. The primary task of the models was to reconstruct and denoise subsurface fractured images, and following reconstruction, a three-class segmentation process (matrix, open fracture, and filled fracture) was performed using the classical Otsu thresholding method to evaluate the improved performance of the reconstructed outputs. Quantitative and qualitative results demonstrated that the Vision Transformer outperformed the autoencoder due to its use of the attention mechanism. The Vision Transformer achieved a peak signal-to-noise ratio of 45 and a structural similarity index of 0.98 in image reconstruction, whereas the autoencoder achieved values of 39 and 0.93, respectively. In the Otsu-based three-class segmentation, the Vision Transformer again delivered superior performance in terms of precision, recall, and F1-score. For the matrix class, precision, recall, and F1-score reached 0.989, 0.990, and 0.989, compared to 0.980, 0.969, and 0.974 for the autoencoder. For the filled-fracture class, the results were 0.974, 0.953, and 0.963, and for the open-fracture class, 0.923, 0.920, and 0.921, while the autoencoder recorded 0.960, 0.962, 0.961 and 0.940, 0.934, 0.937 for the same classes. Overall, the findings indicate that the Vision Transformer is a more efficient and accurate option for CT-scan image reconstruction of fractured rocks, offering significant potential to improve numerical analyses of rock properties such as porosity and permeability in reservoir engineering studies. https://pr.ripi.ir/article_1562.html Iranian oil fields need more than 110 million cubic meters of natural gas to increase daily recovery. Due to the country's severe imbalance in natural gas, the supply of this gas by national oil and gas companies is facing a serious challenge. An alternative option is to use non-hydrocarbon gases such as nitrogen, carbon dioxide, and Flue gas (Combustion gas). Combustion gas is a more attractive option due to its low price and availability. In previous studies, the importance of replacing natural gas and its functional and economic aspects from the perspective of reservoir engineering and increasing the recovery factor were examined.From a subsurface process perspective, the challenge of this gas for injection is the presence of high amounts of water vapor and moderate oxygen concentrations in some incomplete combustion processes, which cause corrosion and reduce the safety of the facility. Continuing previous research, the aim of this study is to design a power plant combustion gas purification unit to separate water vapor and define complete combustion processes for injection into the well. In this designed unit, raw flue gas will be converted into standard gas compatible with the reservoir and can be injected into the field through a precise engineering process. The main objective of this project is to design an optimal process unit for compression and purification and initial calculations of the transmission line. Also, different scenarios for implementing the process have been examined and compared with each other. Finally, the total cost of collecting, purifying, and transporting combustion gas has been calculated for forty discharges of 1, 4, 8, and 20 million cubic meters per day. This study is one of the first studies to document the basic design of standard combustion gas production that can be injected into the field. The design dimensions, the amount of electricity consumed by the facility, and the final cost of the process are presented in four different scales. https://pr.ripi.ir/article_1561.html In this study, the industrial production of ethylene oxide (EO) in a fixed-bed reactor was modeled and simulated using Aspen Plus V14. The catalytic oxidation of ethylene over a silver (Ag) bed requires accurate modeling due to its highly exothermic nature and the presence of side reactions. Reactor performance was evaluated under both steady-state and dynamic conditions. Thermodynamic modeling was carried out using the Soave–Redlich–Kwong (SRK) equation of state and nonlinear kinetic relations, while the effects of pressure drop (calculated using the Ergun equation), heat transfer, and catalyst deactivation were incorporated into the model. Dynamic simulation was performed over a 1100-day period, and the obtained results were compared with actual industrial data, showing good agreement. The influence of ethylene dichloride (EDC) on selectivity, with respect to acceptable EO production levels, was also investigated. Finally, optimization using artificial neural networks (ANN) in the MATLAB environment demonstrated the potential of the proposed model for improving industrial processes. https://pr.ripi.ir/article_1560.html Biodesulfurization is an advanced and efficient approach for eliminating resistant sulfur compounds from fossil fuels using specific microorganisms. This study focuses on the modeling and simulation of key aspects of the biodesulfurization process, including bacterial growth and death trends, substrate consumption rate, and product formation rate. To this end, exponential and logistic kinetic models were combined with ten specific bacterial growth rate models (Monod, Haldane, Aiba, Hinshelwood, Edward, Yano, Blackman, Andrews, Moser, and Webb( to characterize the biomass growth and death. The Luedeking-Piret model was utilized to describe substrate consumption and product formation rates. Accordingly, twenty modeling scenarios were developed, including combinations of growth/death kinetics, specific growth rates, substrate consumption, and product formation. These scenarios were evaluated and validated against experimental data from the literature on dibenzothiophene (DBT) degradation (as substrate) by a Mycobacterium strain, yielding 2-hydroxybiphenyl (2-HBP) as the final product. Statistical analysis shows that the exponential package predictions based on the Webb model, with SSEtotal = 0.087, RMSEtotal = 0.16, and R2mean = 0.976, provide the best fit and the lowest error when compared to the experimental data. This scenario can be applied to design and operate biodegradation units for resistant sulfur compounds from fossil fuels. https://pr.ripi.ir/article_1567.html Finding the best locations to drill in hydrocarbon reservoirs is challenging, particularly in areas with complex geology. To reduce drilling risks and improve productivity, it is important to accurately understand the reservoir’s quality and geomechanical properties. In this study, 3D seismic data were combined with logs from three wells to reduce uncertainty and identify optimal drilling locations in the Nahr-Umr Formation. Geomechanical properties were calculated using pre-stack inversion and then analyzed. These values were normalized using the min-max method and merged into a single petro-geomechanical index (TPG). This index helped create a map showing reservoir quality. A separate attribute was also developed to estimate hydrocarbon potential. This was combined with a reservoir thickness map and a 3D structural model to build a complete system for choosing drilling locations. The best drilling spot was found at the top of an anticline, where reservoir quality is good, hydrocarbon potential is high, and mechanical conditions are stable. A review of the field’s average geomechanical properties showed that, although the Poisson’s ratio is low (indicating brittle rock), the high Young’s modulus, medium density, and good compressive strength help keep the wellbore stable when proper drilling practices are followed. https://pr.ripi.ir/article_1568.html The amount of methane stored in global natural hydrate resources is estimated to be more than twice the total global conventional oil and gas reserves, however recovering energy from this potential resource has proven to be very challenging. The fundamental challenges of gas hydrate dissociation through various approaches are not yet fully understood. In this study, methane gas recovery, water production characteristics and thermal responses of the system under a depressurization approach of 4.6 MPa for a sample of hydrate-containing siliceous sediments with targeted saturation (SH:33%, SA:64%, SG:3%) and methane conversion fraction of 89.2% in a 1-liter crystallizer and at a formation temperature of 279.2 K were investigated. In this study, we found that energy recovery under limited depressurization resulted in slow and incomplete methane hydrate dissociation and intermittent water and gas production behavior. The findings of this study showed that the kinetics of methane hydrate dissociation in low-temperature sedimentary environments with a limited depressurization approach can be affected by stability conditions, which will also lead to the possibility of disruption in the propagation of the depressurization front. Due to the possibility of separating fluids resulting from hydrate dissociation within the crystallizer during depressurization, significantly less water was produced in the first ten hours. The study of the kinetics of methane hydrate formation and dissociation in this study can provide valuable information for analyzing long-term economic production and assessing the life cycle of gas production from hydrate reservoirs. https://pr.ripi.ir/article_1569.html Iran receives 400 billion cubic meters of precipitation annually, 70% of which is lost due to surface evaporation or runoff. Of the approximately 90 billion cubic meters of water consumed by the country, more than 50% is supplied by groundwater resources. Therefore, managing water supply through groundwater reservoirs (tables) is very important. In artificial recharge methods with a flood spreading approach, they often become evaporation ponds after a period of 3 to 5 years due to the high volume of sediments and, in practice, instead of water infiltration, they cause evaporation and water waste. The method proposed in this article is a solution to overcome the two issues of flooding and surface evaporation in water supply. In this method, during the rainy season, river water is added to the aquifer through injection wells after the containment and purification process, and during the hot season, it will be possible to exploit it through production wells.In this study, the technological background was studied along with a case study of process design in one of the aquifers in the country. In addition to preparing a reliable computer model of reservoir behavior and water hydrodynamics in this porous environment, the effect of injection on increasing the water level in the aquifer was studied for 8 injection scenarios from 1 to 20 million cubic meters. Using Prosper software, the pressure drop in the well column and the injection flow rate were studied. The results of the technical section showed that in this aquifer, there is a daily injection capacity of 10 thousand barrels of water for the defined wells without the need for additional pressure.Also, an economic evaluation of the investment amount, final cost and project revenues was estimated. The results show that this technology has the potential to increase storage efficiency in the country's watershed management plans. Also, the final cost of water storage and supply is much more economical than the method of dam construction, desalination and water transfer. https://pr.ripi.ir/article_1570.html In many recent studies, optimizing the concentration and type of ions in injection water has led to effective improvements in oil recovery. In this research, the effect of crude oil type and different aqueous solution compositions on the interfacial properties and critical micelle concentration (CMC) of the cationic surfactant CTAB was experimentally investigated. Two crude oil samples with API gravity of 27.3 and 32.3, differing in asphaltene and resin ratios and with slight differences in total acid and base numbers, as well as in hydrocarbon phase compositions (saturates and aromatics), were used. The experiments included measuring dynamic and equilibrium interfacial tension (IFT) in the presence of deionized water (DW), seawater (SW), and engineered Persian Gulf seawater with the doubled concentration of divalent ions (SO₄²⁻, Ca²⁺, Mg²⁺) (ITW). Although the IFT for both crude oils in the presence of salt reached below 0.1 mN/m, the CMC analysis showed that for the lighter oil (32.3 API) in the presence of ITW, the CMC value decreased to 75 ppm, while for the heavier oil (27.3 API), it remained almost constant at 200 ppm. The lighter oil, with more resin and less asphaltene (higher asphaltene/resin ratio and lower colloidal instability index), provided more favorable conditions for CMC reduction and micelle formation. Adsorption kinetics analysis also showed that the adsorption time (τ) for the light oil in the presence of ITW is shorter, indicating faster surface process kinetics in this system. Overall, the results demonstrated that the synergy between the active ions in smart water and the CTAB surfactant can effectively reduce IFT and CMC, and improve conditions for advanced oil recovery. The findings of this study clearly show that the resin-to-asphaltene ratio and the colloidal instability index play a key and decisive role in the surface activity of brine-surfactant systems. https://pr.ripi.ir/article_1571.html The Fahliyan Formation is recognized as one of the major hydrocarbon reservoirs in the Zagros Basin and the Persian Gulf, mainly composed of carbonate sequences and exhibiting considerable complexity in terms of reservoir characteristics. Rock typing is a key step in petrophysical and geological reservoir studies, as it contributes to a better understanding of reservoir behavior, quality, and fluid flow prediction. In this study, an integrated approach using the Flow Zone Indicator (FZI) method, electrofacies clustering, and microscopic facies analysis has yielded reliable results for rock type determination. In the first step, six microfacies belonging to a ramp setting (including lagoon and shoal deposits) were identified from thin sections of well A2. Subsequently, using the FZI method, four hydraulic flow units were defined based on porosity and permeability data from the Fahliyan reservoir in well A2. By integrating petrographic and petrophysical data, reservoir quality was ranked from very poor (RT1) to good (RT4) based on the four identified flow units. Then, using data from three petrophysical logs (effective porosity, neutron, and sonic), electrofacies of the Fahliyan Formation were determined using the MRGC clustering method, resulting in the identification of 11 electrofacies. The combination of sedimentary facies, electrofacies, and rock typing indicates that the blue electrofacies (RT4, MF6) represents good reservoir quality, whereas the red electrofacies (RT1, MF1, MF2) represents very poor reservoir quality. This integrated approach provides an effective framework for zoning and ranking reservoir quality in heterogeneous carbonate reservoirs such as the Fahliyan Formation. https://pr.ripi.ir/article_1572.html Accurate water-cut forecasting is essential in mature water-drive reservoirs, where excessive water production is the primary cause of well shutdowns and reduced field productivity. This study develops an advanced deep-learning framework using Long Short-Term Memory (LSTM) networks to predict water cut in a giant Iranian oil field characterized by strong aquifer support, heterogeneous lithology, and multi-layer completions. A high-resolution dataset comprising 33,697 water-cut measurements was comprehensively preprocessed through duplicate removal, outlier filtering, feature-variance analysis, technical consistency checks, smoothing, and missing-value imputation. After Monte Carlo sensitivity analysis, seventeen static and dynamic features—such as well coordinates, perforation geometry, permeability, oil-rate history, and production time—were selected as model inputs.A time-consistent data split was applied, where 80% of early-time data were used for training and the remaining 20% for testing to avoid information leakage. The final LSTM architecture, optimized using grid-search hyperparameter tuning, was trained independently for four reservoir zones. The model achieved excellent performance, with Pearson correlation coefficients of 0.957–0.976 on training data and 0.889–0.920 on testing data, while Absolute Relative Error distributions remained tightly centered near zero. The LSTM successfully reproduced long-term production trends, short-term fluctuations, and operational interventions such as recompletions and flow-rate modifications. These results demonstrate that the proposed LSTM-based framework provides a robust and scalable tool for water-cut prediction and supports informed reservoir-management decisions in large heterogeneous water-drive fields https://pr.ripi.ir/article_1573.html Elemental sulfur deposition represents a major operational challenge in sour gas reservoirs and in gas production, transportation, and processing facilities. A decrease in sulfur solubility under operating conditions can lead to flowline blockage, production losses, intensified corrosion, and serious flow-assurance problems. Therefore, accurately predicting sulfur solubility across varying temperatures, pressures, and gas compositions is of substantial industrial relevance. In this study, molecular dynamics (MD) simulations were employed using the LAMMPS package to predict the solubility of elemental sulfur in eight sour-gas mixtures, including six binary and two ternary systems. The molecular modeling framework utilized validated force fields: the Ballone potential for S₈, a three-center model for CO₂, and Lennard-Jones and OPLS-AA models for the remaining gas components. Simulations were carried out in the NPT ensemble for 21 ns, and equilibrium states were identified after removing the transient region based on potential energy, kinetic energy, and the number of free S₈ molecules. Sulfur solubility was quantified by determining the number of S₈ molecules dissociated from the solid phase and dispersed into the gas phase using cluster-analysis techniques, followed by conversion into engineering units. The predicted values were compared against reliable experimental data for all gas mixtures. The results demonstrated that the MD approach accurately reproduced sulfur solubility, with an overall relative error below 1%. Sensitivity analysis indicated that selecting an appropriate cutoff radius significantly enhances numerical stability and model accuracy. Overall, the findings demonstrate that molecular dynamics is capable of reliably predicting the equilibrium solubility behavior of elemental sulfur in binary and ternary sour-gas mixtures. This computational approach provides a robust tool for analyzing sulfur deposition in sour-gas reservoirs and for optimizing the design and operational conditions of processing and transportation systems. https://pr.ripi.ir/article_1574.html Silver nanoparticles hold a special place in various branches of nanotechnology due to their high antifungal efficacy and biological stability. The use of green methods, such as synthesis with plant extracts and microwave heating, can simultaneously improve both the performance and biocompatibility. Optimizing the production conditions leads to a balance between physical characteristics and biological effects. In the present study, mulberry leaf extract was used as a reducing agent for the green synthesis of silver nanoparticles. Using response surface methodology (RSM) for experimental design, the synthesis process was optimized within an extract volume range of 5 to 15 mL and a microwave heating duration of 1 to 5 minutes. The optimization results indicated that an extract volume of 11.39 mL and a heating time of 8.6 minutes were the most suitable conditions for the synthesis process. Under these optimal conditions, the highest concentration and the smallest average nanoparticle size were achieved at 67.11 ppm and 66 nm, respectively. Finally, evaluation of the nanoparticles synthesized under the optimal conditions revealed antioxidant and antifungal activities of 52.31% and 74.17%, respectively. https://pr.ripi.ir/article_1575.html Coiled tubing (CT) units operate at higher speeds compared to conventional drilling rigs, while offering lower operational costs and more compact equipment. Additionally, they require less time for rig-up and provide greater overall efficiency. Consequently, CT technology has found extensive application in the oil and gas industry. Nevertheless, CT is susceptible to fatigue failure due to cyclic bending stresses experienced on the reel and guide arch, particularly under high-pressure operating conditions. The fatigue life of CT can be evaluated through three main approaches: the running foot method, the laboratory (trip) method, and the theoretical method. However, due to the inherent limitations of the running foot and laboratory methods, the development and application of predictive fatigue life models have become increasingly important. The present study aims to estimate the fatigue life of CT using artificial intelligence (AI)–based modeling techniques. For this purpose, 249 datasets were compiled from previously published studies and research articles. Three distinct machine learning algorithms—Support Vector Regression (SVR), Random Forest, and CatBoost—were employed to develop predictive models for CT fatigue life as a function of the number of trips. The primary objective was to identify the most effective algorithm for accurate fatigue life prediction. The results obtained from evaluating the three machine learning models indicated that their performance on the test dataset was very close to each other. This finding demonstrates the convergence of the models in learning the data patterns and the proper consistency between the input variables and the target output. Among these, the SVR model achieved the highest R² and the lowest AAPE and RMSE values, demonstrating superior performance. The other two models also exhibited satisfactory accuracy. https://pr.ripi.ir/article_1576.html Underground Hydrogen Storage (UHS) has gained increased importance as a key solution for future green energy economy. However, the equilibrium of hydrogen with mineral-affinitive microorganisms present in underground reservoirs can lead to hydrogen consumption and methane production through the hydrogenotrophic methanogenesis process. This phenomenon not only reduces the amount of stored hydrogen but also affects the safety and economy of this strategy. In this study, geochemical and biological reactions related to hydrogen consumption in underground environments were simulated using PHREEQC software. The model was based on operational and kinetic parameter ranges reported in previous studies, along with the evaluation of methanogenic reactions. Then, 27 different scenarios were simulated in the form of a three-factor experimental design (temperature: 40, 55, and 70°C; pressure: 1500, 2000, and 2500 psi; reservoir rock type: quartz, calcite, and dolomite). The results showed that the least hydrogen consumption and methane production occur at higher temperatures (70°C), higher pressures (2500 psi), and in reservoirs formed from quartz rock. This behavior is justified by the chemical inactivity of the quartz surface and the lack of providing a suitable environment for microbial growth. In contrast, carbonate rocks, especially calcite, due to higher solubility and easier release of metal ions (such as Ca²⁺), create a favorable environment for the enzymatic activity of methanogens. The goal of this project was to identify optimal conditions for hydrogen storage with minimum losses.