Mari Petroleum

Abstract This paper presents a feedforward artificial neural network (ANN) workflow for predicting missing shear sonic logs (DTS) in Field A of the Central Indus Basin using routinely acquired wireline data. The objective was to reconstruct DTS in wells where shear sonic measurements were not acquired because of operational or economic constraints and where empirical correlations did not provide sufficiently consistent estimates. Gamma ray, compressional slowness, neutron porosity, and deep resistivity were used as input logs after quality control, screening of anomalous values, and normalization. The model was trained on four wells containing measured DTS and subsequently applied to 18 wells in which DTS was unavailable. Model performance was evaluated by comparing measured and predicted DTS responses in the calibration wells and by constructing cross-plots between the two datasets. The predicted and measured DTS showed strong agreement, with a correlation coefficient of 0.88. After calibration, the trained network generated continuous DTS curves in the remaining wells and extended shear-sonic coverage across the Lower Goru B-Sands interval. The reconstructed curves preserved the principal vertical trends observed in the measured data and provided a consistent basis for subsequent petrophysical and geomechanical interpretation. The results indicate that ANN-based DTS prediction can provide a practical alternative in fields where direct shear sonic acquisition is limited. The generated logs can support reservoir characterization, seismic-to-well integration, and geomechanical workflows that depend on shear-sonic information. The proposed workflow is therefore applicable to data-limited fields with comparable geological conditions and log availability.

The petroleum system in the Risha gas field, eastern Jordan, is not fully understood. The current study aims to evaluate source rocks, reservoirs, and seals within the basin by integrating 1D basin models and well logs together with geochemical and petrophysical data. Two effective source rocks are identified within the Paleozoic shale intervals, namely the Mudawwara and Dubeidib formations. The Mudawwara and Dubeidib formations with TOC values ranging from 0.51 to 6.75 wt%, which suggests poor to very good potential. The formations contain primarily Type III (gas-prone) and mixed Type II/III (oil/gas-prone) kerogen. The thickness of the Paleozoic sedimentary section increases to the east. In contrast, the thickness of the Mesozoic section increases to the west, and the entrapment area in the middle to the eastern part of the study area is characterized by the accumulation of petroleum in the Ordovician sandstone reservoirs. The Risha sandstone reservoir has an increased net pay thickness in east and southeast directions. The migration of hydrocarbons petroleum in the Risha gas field is strongly suggested as a direct primary migration from the source rock intervals within the Mudawwara and Dubeidib shales to the intercalated sandstone reservoirs of the Risah and Dubeidib formations. In addition, vertical migration along faults and up-dip migration to the middle and eastern parts of the study area are also suggested. The Risha-14 well represents the ideal petroleum system in eastern Jordan, as it encompasses all the key elements of petroleum generation and accumulation. The critical moment for the source rock is estimated to have occurred approximately 250 Ma ago. The effective shale source rocks of the Mudawwara Formation were deposited in the Early Silurian. They entered the main oil window in the Early Devonian (similar to 410 Ma ago), with significant gas generation occurring in the early Carboniferous (similar to 330 Ma ago). The Dubeidib Formation sandstone reservoir was deposited during the Ordovician (485-458 Ma ago). Structural traps formed during the Paleozoic (280-250 Ma ago), formed during the Hercynian Orogeny through rifting and compression phases were subsequently charged with petroleum through migration and accumulation.

Recent research has identified a key limitation in the petroleum sector commonly used pour point depressant (i.e., EVA-copolymer), which shows decreasing efficiency as the paraffin wax content increases. To address this issue, this study explored the enhancement of EVA-copolymers as pour point depressants for waxy crude oil through the gamma irradiation grafting of homopolymers with long alkyl chains onto EVA copolymers to produce (EVA-g-p(HDOAB) and EVA-g-p(HDAOCP)+Br−). FTIR and 1H-NMR spectroscopy show that EVA-g-p(HDOAB) and EVA-g-p(HDAOCP)+Br−) were successfully prepared using gamma radiation. Various concentrations (100, 500, 1000, 2000, and 5000 ppm) of the grafted EVA-copolymers were introduced into the crude oil. The grafted copolymers showed a significant reduction in pour point from 24°C to -18°C and -15°C at a 2000 ppm concentration, outperforming pure EVA, which only reduced the pour point to -9°C. Additionally, both grafted copolymers demonstrated notable viscosity and shear stress reductions, particularly at temperatures close to the pour point. The enhanced low-temperature flow properties are attributed to the synergistic effect of the grafted EVA-copolymer acting as nucleating agents that modify crystallization patterns and inhibit crystal growth. This study underscores the potential of using gamma ionizing radiation to develop high-performance PPDs for crude oil.

Accurate estimation of formation conditions plays a pivotal role in effectively managing various processes related to hydrates, including flow assurance, deep-water drilling, and hydrate-based technology development. The formation temperature of methane hydrates in the presence of brine greatly affects the efficacy and accuracy of these processes. This work presents a comprehensive and novel comparative analysis of nine distinct machine learning models for accurate prediction of formation temperatures of methane hydrate. This study investigated the application of major machine learning (ML) algorithms including multiple linear regression (MLR), long short-term memory (LSTM), radial basis function (RBF), support vector machine (SVM), artificial neural network (ANN), gradient boosting regression (GBR), gradient process regression (GPR), random forest (RF), and K-nearest neighbor (KNN). The model accuracy was validated against a large dataset comprising of over 1000 data points with diverse range of salt concentrations. In this regard, model accuracies were compared using several metrics including R 2, ARD, and AARD. The experimental results exhibited KNN algorithm to be fast-converging, accurate, and consistent over the entire range of data points with an R 2 score of 0.975 and AARD of 0.385%. The results enable efficient and accurate temperature estimation with ML algorithms for multiple hydrate-related processes.

Chemical flooding is a crucial technique in petroleum recovery. Although synthetic polyacrylamides are widely used, they suffer from hard reservoir conditions (high salinity, temperature, and pressure) and high costs. Current efforts focus on eco-friendly and affordable biopolymers like xanthan gum to overcome these issues. This study screens xanthan gum modification to improve its rheological properties and tolerance to high temperature, salinity, and shearing action by copolymerizing it with vinyl silane, vinyl monomers, and silica nanoparticles. The new composite was characterized using Fourier Transform Infrared Spectroscopy (FTIR), Thermal Gravimetric Analysis (TGA), Atomic Force Microscopy (AFM), and proton Nuclear Magnetic Resonance (NMR) tests. Its implementation was evaluated in polymer flooding at 2200 psi pressure, 135,000 ppm salinity, and 196°F temperature. Unlike previous studies that evaluated xanthan gum at 176 °F, 1800 psi, and 30,000 ppm, without combining those three factors in one experiment. The rheological properties of native and composite xanthan were examined at reservoir conditions, as well as their viscoelastic properties (G′ and G″). Flooding runs used actual Bahariya formation cores at the lab scale. Simulation studies were conducted on a lab/field scale using the tNavigator simulator and economic feasibility to calculate the net present value. The most outcoming findings of this research comprise (1) investigating the impact of salinity, temperature, and pressure on the rheological properties of native and composite xanthan. (2) The composite xanthan exhibits more resistant criteria, as it recovered 27% residual oil versus 22% for native xanthan. (3) Modeling and simulation studies exhibit 48% oil recovery for composite versus 39% for native xanthan and 37% for water flooding. (4) Economically, using native and composite xanthan through enhanced oil recovery methods increased net present value to $32 mm and $58 mm versus traditional methods.