В данном исследовании представлен устойчивый и адаптивный подход к переработке полезных ископаемых. Была разработана гибридная интеллектуальная система управления для обогащения мелкодисперсной хромитовой руды в отсадочной машине. Цель состоит в повышении эффективности разделения и снижении потерь хрома за счет оптимизации параметров процесса в реальном времени при переменных условиях подачи. Метод учитывает колебания состава руды путем интеграции трех компонентов: физического моделирования движения частиц, регрессионного анализа и прогнозирования на основе нейронной сети. Уровень слоя отсадочной машины и частота пульсаций используются в качестве управляющих переменных, а содержание Cr₂O₃ в подаваемом сырье (формула представлена) рассматривается как возмущение. Нейронная сеть прогнозирует содержание Cr₂O₃ в концентрате ( формула представлена) и в отходах (формула представлена)), представляющих собой , соответственно, богатую хромитом и пустую фракцию. Оптимизация выполняется с использованием алгоритма внутренней точки с ограничениями. Модель демонстрирует высокую точность прогнозирования со среднеквадратичной ошибкой (MSE) ниже 0,01.

В данном исследовании предлагается интеллектуальный алгоритм управления для промышленных процессов с множественным входом и множественным выходом (MIMO). Этот алгоритм основан на интеграции цифрового двойника (DT), модельного прогнозирующего управления (MPC), генетического алгоритма (GA) и нейронной сети (NN). Разработанная архитектура использует гибридную схему MPC, включающую дополнительную ветвь коррекции на основе нейронной сети. Рабочий процесс включает предварительную обработку входных данных, линеаризацию рабочей точки и обучение нейронной сети, вычисление оптимальной последовательности управления на скользящем горизонте, управление с обратной связью и адаптацию на основе ошибки прогнозирования. Этот инновационный гибридный закон управления использует линейную модель пространства состояний в качестве базового предиктора и компактную надстройку нейронной сети для компенсации неучтенных нелинейностей. Генетический алгоритм ищет оптимальную последовательность управляющих воздействий, соблюдая ограничения процесса и обеспечивая стабильное использование коррекции на основе нейронной сети.

The Shu-Sarysu Basin in central-southern Kazakhstan remains one of the underexplored gas-prone provinces, with 12 discovered gas fields including Amangeldy (884 Bcf) and Pridorozhnoye (225 Bcf). In the context of global energy transition, such basins require integrated geological assessment to constrain exploration potential. Historical studies within the region were spatially limited and prematurely discontinued, resulting in fragmented datasets and a lack of modern interpretation. This review reassesses published geological data within a petroleum systems framework, applying contemporary geodynamic and stratigraphic concepts. Analysis shows that tectonostratigraphic evolution of the basin during Devonian–Permian time (390–250 Ma) favored formation of mature, gas-prone systems within structurally compartmentalized troughs, with effective source, reservoir, and seal configurations. Building on these findings, a three-tier classification of exploration zones is proposed based on system maturity, trap integrity, and gas shows, reflecting geological success probability. This provides a basis for prioritizing future exploration despite limited seismic and drilling coverage in many areas. Recommended priorities include digitization of archival data, structural modeling, modern geochemical and diagenetic analysis, and focused evaluation of promising areas to support future exploration.

The Devonian–Permian succession of the Tasbulak Trough in the Shu–Sarysu Basin contains confirmed gas shows (wells 462, 1-P Izykyr, 1-P Sokyr-Tobe, and 1-P Kamenistaya) and a sedimentary cover exceeding 5500 m but still lacks a unified 3D structural interpretation capable of explaining the distribution of gas-prone intervals. This study addresses this gap by digitizing and integrating legacy well and 2D seismic datasets to construct horizon-consistent three-dimensional structural surfaces for eight target horizons. The resulting model reveals a low-deformation structural framework dominated by a previously undocumented element—the Central Tasbulak Ridge—which exerts first-order control on fault segmentation, trap geometry, and gas preservation. Structural surfaces were synthesized with stratigraphic intervals to define reservoir–seal–trap relationships, highlighting the late Visean–early Serpukhovian carbonate subformation as the primary target interval. Building on these relationships, a prospect evaluation matrix was developed to classify structural, stratigraphic (including intraformational), and combination trap types together with their corresponding sealing units. The results demonstrate long-term tectonic stability, multi-level evaporitic seals, and inheritance-guided trap evolution, providing a reference framework for assessing gas prospectivity in data-limited intracratonic basins and advancing understanding of petroleum-system architecture in stable continental settings

Freeze-drying of camel milk is time-consuming due to its high density, protein, and fat content, which limits industrial application and increases energy demand. In this study, ultrasonic treatment (UT) and homogenization were applied as pretreatments prior to freeze-drying. A full factorial design (24) was used to evaluate process factors. Experimental trials were combined with mathematical modeling based on heat and mass transfer equations (Fourier’s law, Fick’s law, and dimensionless criteria Re, Pr, Nu). Pretreatment significantly improved drying efficiency: the drying time of camel milk was reduced from 28 h (control) to 22 h (treated samples), corresponding to a 21% decrease. Protein content and solubility were enhanced, while residual moisture decreased to 3–4%. The developed equations accurately predicted sublimation rates and drying kinetics, with R2 = 0.95 confirming high model reliability. The integration of ultrasonic treatment and homogenization into freeze-drying provides a viable strategy for reducing drying time and improving product quality of camel milk. The established criterion equations and models support process optimization and industrial application.

Vacuum freeze-drying is a preferred method for converting mare’s milk into a shelf-stable powder while retaining its nutritional and sensory qualities. However, the process remains time-consuming and energy-intensive. This study presents a validated three-dimensional numerical model that simulates conjugate heat- and mass-transfer during vacuum freeze-drying under industrially relevant conditions. The governing equations for heat conduction, water-vapour transport (Darcy’s law), and latent-heat removal were solved in COMSOL Multiphysics 6.3 using the Deformed Geometry interface to track phase-front motion without remeshing. Simulations were performed on a 50 × 45 × 0.7 cm slab frozen to -50 ℃ with a shelf temperature of -20 ℃ and chamber pressure ranging from 15 to 35 Pa. Validation was conducted in a 0.45 m2 pilot dryer using centre-line temperature measurements and gravimetric moisture-loss data. The model reproduced experimental temperature profiles (R2 = 0.96) and predicted sublimation-front motion within 4% of image-based observations. Thinner samples (3 mm vs. 7 mm) and lower pressures (15 Pa vs. 35 Pa) significantly reduced drying time, though gains diminished below 15 Pa. This model offers a practical framework for optimising freeze-drying conditions and is extendable to other porous food and biopharmaceutical products. © The Author(s) 2025. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits the use, sharing, adaptation, distribution and reproduction in any medium or format, as long as appropriate credit to the original author(s) and the source is given by providing a link to the Creative Commons license and changes need to be indicated if there are any. The images or other third-party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The article presents the results of calculations of sublimation drying of mare's milk without cracks in the porous structure of the product and taking into account the formation of a crack. In the case of a crack, the calculation was carried out taking into account the cracking of the product at 4:00 h and the change in the permeability of the porous body during crack formation. Stefan's problem of the mare's milk sublimation process is solved by the numerical phase field method. The calculations obtained the distributions of temperature, pressure and steam velocity, showing the development of the sublimation front in the porous structure of the product over time. Without a crack, the sublimation front spreads from top to bottom of the product almost uniformly along the length of the porous structure and is in agreement with the known data. When cracks form in the porous structure, the front disintegrates into semi-independent drying zones. The crack leads to a decrease in pressure at the sublimation front and draws in steam from the product in its vicinity.

The aqueous solution of copper (II) complex of poly(vinyl alcohol) (PVA-Cu(II)) was characterized by conductimetric titration, UV-Vis spectroscopy and FTIR. The molar composition of the PVA-Cu(II) complex was determined to be equimolar. Subsequently, the PVA-Cu(II) complex was reduced by sodium borohydride to prepare copper nanoparticles (CuNPs) stabilized by PVA (PVA-CuNPs). According to DLS measurements in aqueous solution the average size of PVA-CuNPs was varied from 10 to 25 nm depending on the amount of sodium borohydride used for reduction. The optimal volume of sodium borohydride to obtain 96 % PVA-CuNPs with 10 nm size was found to be 3 mL of NaBH4 (0.5 mol⋅L–1). The PVA-CuNPs were then deposited onto SiO2 support to obtain SiO2/PVA-CuNPs nanocatalyst for the oxidation of 1-propanol. The SEM image and XRD spectrum of SiO2/PVA-CuNPs nanocatalyst showed the deposition of PVA-CuNPs on the surface of SiO2. The resulting SiO2/PVA-CuNPs nanocatalyst was used for the oxidation of 1-propanol to propionaldehyde by molecular oxygen in a batch-type catalytic reactor at 20oС and atmospheric pressure. The optimum catalyst mass and reaction time were found for the conversion of 1-propanol to propionaldehyde with yields ranging from 61.4 % to 87.8 %. Criteria of hydrodynamic and diffusive similarity (Re, Pr’, Sh), overall volumetric mass transfer coefficient (kLα), and economic metric (STY) were evaluated. 1-Propanol reacted with the decomposed atomically adsorbed oxygen atoms on the Cu(111), (220) surfaces to form propionaldehyde and water.

Gellan gum prepared from the domestic raw materials of Kazakhstan is studied by methods of 1H NMR spectroscopy, FTIR, GPC, TGA, and compared with characteristics of commercial gellan gum. Fermentation on glucose-fructose syrup of Zharkent and Burunday corn starch plants (Kazakhstan) by Sphingomonas paucimobilis ATCC 31461 produces a biomass containing a high acyl gellan gum (HAG). The average molecular weights, Mw, Mn, Mz, and polydispersity index (PDI) of HAG are determined by GPC. Low acyl gellan (LAG) is obtained by treatment of HAG with alkaline solution. It is shown that the spectral and thermal characteristics of HAG and LAG produced from the glucose-fructose syrup and commercial gellan gum are similar.

Commercial low acyl gellan (LAG) is purified by the fractional dissolution method. An aqueous solution of purified commercial LAG exhibits polyelectrolyte character due to its high molar mass and carboxylic groups in glucuronic acid fragments. The empirical Fuoss equation determines the intrinsic viscosity of purified LAG in salt-less water. Fractionation of commercial LAG is carried out by ultrasound treatment. LAG fractions are characterized by 1H NMR and FTIR spectroscopy. The intrinsic viscosities of ultrasonically treated LAG samples are measured in 0.025 m of tetramethylammonium chloride (TMACl). The molecular weights of LAG fractions are determined by the Mark–Kuhn–Houwink equation.