Atyrau University of Oil and Gas

This paper addresses the problem of multi-class classification of bacterial microscopic images using a rigorous experimental protocol designed to prevent information leakage and improve performance. The dataset consists of 2034 images representing 33 taxa, organized by class. Data integrity checks confirmed the absence of corrupted or unreadable files. To formalize image characteristics and ensure quality control, indirect geometric and textural features were calculated, including minimum frame size, brightness statistics (mean and standard deviation), Shannon entropy, Laplace variance, and Sobel gradient energy. Quality checks revealed a small proportion of images with extreme brightness (2.5074%), while no samples with critically low sharpness according to the selected criteria were detected. Statistical analysis of interclass differences using the Kruskal–Wallis test with multiple comparison correction demonstrated the high discriminatory power of texture features, specifically gradient energy (ε2 = 0.819987) and Laplace variance (ε2 = 0.709904). Feature correlations were consistent with their physical interpretation, revealing a strong positive relationship between sharpness and gradient energy. Principal component analysis confirmed a strong structural pattern, with the first two components explaining 75.5766% of the total variance. For a unified comparison, classical machine learning, transfer learning, and modern deep architectures were evaluated within a single protocol.

Abstract Precise forecasting of oil production rates is essential for effective reservoir management and economic decision-making in the petroleum industry. Traditional techniques such as Decline Curve Analysis and Numerical Reservoir Simulation are limited in capturing the complexity of unconventional reservoirs. This study presents a framework that applies two machine learning models: Random Forest (RF) and Long Short-Term Memory (LSTM). The models were tested on production data from a large oil field in Kazakhstan for the period 2003 to 2020. The RF model identified Water Cut as the main factor, while the LSTM model captured time-dependent patterns in the data. The R2 and MSE results for the training set were 0.98 and 281 for RF, 0.93 and 807 for LSTM, and 0.95 and 512 for the Hybrid model. When it comes to the testing dataset, the results for error metrics were 0.83 and 2769 for RF, 0.94 and 899 for LSTM, and 0.88 and 53 for the Hybrid model. These results were further supported by the model's selection criteria: the Hybrid was characterized by the minimum AIC and BIC, which demonstrates its better balance between forecasting accuracy and model parsimony. To the best of our knowledge, this is among the first attempts to utilize such a hybrid model for oil production predictions in Kazakhstan. The findings demonstrate the potential of this tool as a valuable instrument in reservoir management and production planning.

One of the urgent tasks is to improve the environmental characteristics of fuel and reduce harmful emissions into the atmosphere using oxygenates to increase the octane number of gasoline compositions. This article aims to study the effectiveness of dimethyl ethynylcarbinol (DMEC) as a component for increasing the octane number of compositions consisting of straight-run gasoline (SRG) and reforming gasoline (RG). Condensation of acetone with acetylene by Favorsky reaction conditions, express analysis on SHATOX octanometer, and preparation of gasoline test samples with added DMEC to achieve the result were utilized. This article presents for the first time the results of investigation of the effect of methyl tert-butyl ether (MTBE) and DMEC on increasing the octane number of gasoline blends. A mixture of SRG with RG in the ratios of 40:60 and 20:80 was utilized to create gasoline compositions. The effect of the investigated oxygenates on the octane number of gasoline compositions was evaluated. The results demonstrated that DMEC has a higher effect in increasing the octane number than methyl-tert-butyl ether. The practical significance of the study is that the results can be utilized for improving gasoline production technology and creating a more environmentally friendly and efficient fuel.

This study analyzes the effect of artificial aging on the mechanical deformational and optical properties of various paper samples, which allows us to evaluate their durability and suitability for long-term storage. The methods of accelerated aging, measuring the breaking length, specific resistance, elongation, and fracture strength, were used, and the optical characteristics were estimated by the R457 and CIE whiteness indices, as well as opacity. Mechanical measurements (breaking length, specific resistance, elongation, and fracture strength) revealed that bleaching reduces residual lignin and strengthens interfiber bonds, boosting pine pulp strength by up to 8%. Optical properties initially improve slightly, then increase sharply after the second bleaching cycle and stabilize, while opacity decreases, providing greater light transmittance. After accelerated aging, the following deterioration is observed: for bleached samples, R457 whiteness changes; and for unbleached samples, CIE whiteness and opacity increase. After aging, aspen pulps and kraft papers retained over 90% of their initial strength and whiteness, whereas untreated and office papers lost up to 20–25%. These findings identify that aspen-based and kraft papers demonstrate better mechanical deformational and optical properties, which makes it possible to predict the operational characteristics of paper depending on the processing and aging methods used.

The research investigates the thermal behavior of mixed systems based on natural and artificial cellulose fibers used as precursors for carbon nonwoven materials. Flax and hemp fibers were employed as natural components; they were first chemically treated to remove impurities and enriched with alpha-cellulose. The structure, chemical composition, and mechanical properties of both natural and viscose fibers were studied. It was shown that fiber properties depend on the fiber production process history; natural fibers are characterized by a high content of impurities and exhibit high strength characteristics, whereas viscose fibers have greater deformation properties. The thermal behavior of blended compositions was investigated using TGA and DSC methods across a wide range of component ratios. Carbon yield values at 1000 °C were found to be lower for blended systems containing 10-40% by weight of bast fibers, with carbon yield increasing as the quantity of natural fibers increased. Thus, the composition of the cellulose composite affects carbon yield and thermal processes in the system. Using the Kissinger method, data were obtained on the value of the activation energy of thermal decomposition for various cellulose and composite systems. It was found that natural fiber systems have three-times higher activation energy than viscose fiber systems, indicating their greater thermal stability. Blends of natural and artificial fibers combine the benefits of both precursors, enabling the deliberate regulation of thermal behavior and carbon material yield. This approach opens up prospects for the creation of functional carbon materials used in various high-tech areas, including thermal insulation.

This study investigates the impact of CO2-enhanced oil recovery (CO2-EOR) on the petrophysical properties and oil recovery potential of sandstone reservoirs in the oilfields located in the east-southern Precaspian region of Kazakhstan. Despite the recognized potential of CO2-EOR for improving oil recovery and aiding carbon sequestration, there is limited understanding of how CO2-EOR specifically affects the petrophysical properties of sandstone reservoirs in this region. Laboratory experiments were conducted using two core samples from the selected oilfields to examine changes in porosity, permeability, and oil recovery coefficients. The results demonstrated that porosity changes ranged from a slight increase of 1.1% to a decrease of 1.5%, while permeability reduction was significant, with decreases ranging from 29% to 50% due to clay alteration and halite precipitation. The oil recovery coefficient after CO2 flooding was found to be between 0.49 and 0.54. These findings underscore the complex interactions between CO2 and reservoir rocks, emphasizing the need for tailored EOR strategies in different geological settings.

The article discusses research aimed at determining the effectiveness of measures to intensify oil production in the fields of Western Kazakhstan. Based on the results obtained, recommendations were developed for the further application of oil production intensification methods for the conditions of West Kazakhstan fields. The highest oil displacement coefficient (ODC) in the development of high-viscosity oils can be achieved using thermal production methods. To reduce viscosity, in practice, deep warming of the bottom-hole zone of the well is used, which can be carried out by electrothermal treatment of the bottom-hole zone of the formation, steam treatment of wells using downhole solid fuel heat generators, and so on. A combined heating method is proposed that allows heating the well products. Heating is provided by a solar panel all year round. In case of a decrease in solar activity, it automatically switches to a constant electric current. This method differs from others in that the solar panel continuously supplies electricity, reducing the consumption of electrical energy. In addition, the article presents the results of studies of the optimal heating temperature of the bottom-hole formation zone, potential well flow rates at this temperature and oil viscosity. Recommendations are given on the most rational way of heating the bottom-hole zone of wells in the fields of Western Kazakhstan.