The object of the research is meat raw materials obtained from animal slaughter. The research is aimed at addressing the problem of the efficient utilization of meat raw materials under conditions of their natural variability and production uncertainty, which limits the ability to make timely and well-founded technological decisions. The research developed a mathematical model for planning the production of a diverse assortment of meat products, taking into account regional characteristics of raw materials and the production and technological potential of enterprises. The use of the model enables the formation of the consumer properties of meat products, including grade, taste, nutritional and biological value, through the rational utilization of raw materials of different origins, considering the species, breed, age, sex, and fatness of animals. The article shows that product quality should be evaluated using a comprehensive criterion integrating physicochemical, functional-technological, and structural-mechanical indicators. Particular importance is given to rheological characteristics correlating with properties of raw materials, minced systems, and finished products. Within the research, an objective function was developed that accounts for availability and quality of raw materials for different products. Sets of enterprises, assortments, animal characteristics, and supply zones were formalized considering environmental conditions of origin areas. Production constraints were described using a compatibility matrix, emphasizing water-binding processes and preservation of intramuscular fat as key quality factors. Production variants were also formalized, considering technological constraints under conditions of production uncertainty. The practical application of the developed model provides digital support for production management, optimizes raw material allocation, and enables product yield prediction. It also reduces production losses and meets the current requirements for the digital transformation of meat processing enterprises.
The object of research is the recovery of moringa powder in the production of instant functional drinks. Despite the existence of researches on the use of moringa powder in functional products, the relationship between the structural and morphological characteristics of powder particles and the kinetics of its recovery requires in-depth research, which is important for substantiating the technological parameters of the production of instant functional drinks. The microstructure of moringa powder was studied using scanning electron microscopy, and the elemental composition was determined using energy-dispersive X-ray spectroscopy. The powder had a pronounced polydispersity with a predominance of stable cell conglomerates of irregular shape with a size of 5 to 130 μm. The presence of a developed rough surface and regular porous channels (from 5 to 15 μm) provides an increase in the specific area of contact with the liquid. A high protein potential was revealed due to the significant N content (9170 mg/100 g). It was found that consuming 100 g of a drink based on moringa powder provides 17.9% of the daily requirement of K without the risk of excess salt intake (Na covers only 3.3–3.8% of the requirement). Moringa powder is an excellent source of Mg (970 mg/100 g). This confirms its functional ability to contribute to the normalization of the nervous system. In addition, the functional properties of moringa powder and its positive effect on energy metabolism are provided by the high content of P (1100 mg/100 g). Separately, a significant content of Cr (0.36 mg/100 g) was noted, which is an active regulator of carbohydrate metabolism and blood sugar levels. The relationship between the microstructure, mineral composition and hydration behavior of moringa powder has been proven, confirming its use as an ingredient in powdered functional food systems.
The object of research is the corrosion of St20 steel in aqueous electrolytes of different salt compositions under aerated and deoxygenated conditions. The problem addressed is the development and comparative evaluation of corrosion rate prediction models based on a limited experimental dataset, with the possibility of interpreting the influence of individual factors. The effects of exposure time, NaCl, Na2SO4, and Na2CO3 concentrations, and aeration and deoxygenation conditions on the corrosion rate were investigated experimentally. Three models were considered: a physics-informed LSTM model, an LSTM model with an attention mechanism, and a Random Forest model. The training dataset was formed by pseudo-temporal expansion of the experimental data using a sliding window. This approach transforms discrete measurements into sequences for model training without replacing experimental corrosion rate determination with a purely computational procedure. The models generally reproduce the experimental dependencies within the dataset. The R2 values for the physics-informed LSTM model, the LSTM model with the attention mechanism, and the Random Forest model were 0.9926, 0.9925, and approximately 1. The physics-informed LSTM model accounts for differences between aerated and deoxygenated conditions, which are important for describing corrosion processes in aqueous media. The attention mechanism provides model interpretability by estimating the contribution of individual factors to the prediction. The Random Forest model was used for comparison with neural network approaches and to assess the reproduction of dependencies in the prepared experimental data. The results confirm the feasibility of combining experimental data and machine learning methods to model complex nonlinear corrosion processes. The approach can be used for preliminary assessment of the corrosion activity of aqueous media and for planning further laboratory and industrial researches.
The object of research is the process of designing and modeling intelligent agents of a neural network for the behavior of non-player characters (NPCs). Today, the field of video game development continues to progress, but the methodology for creating the “brain” of NPCs still uses primitive approaches with technological limitations. This creates the problem of ludonarrative dissonance, which actualizes this work, that offers a new solution. The research results demonstrate the potential of a new approach to creating intelligent and realistic NPCs in video games. The developed models demonstrate convergence of training, which confirms the ability to understand complex concepts, such as the psychological personality model International Personality Item Pool 50. The conclusions of the work are based on the results of theoretical analysis, technical modeling and practical experimentation. Analysis of the collected data forms the final confirmation of the feasibility of the proposed theory. An increase in the total step reward from –0.00275 to 0.03176 by the median for the best option and from –0.00733 to 0.02734 for the worst was recorded, taking into account the mathematical limit in [–1.0219; 0.9861] and aggressive normalization of the hyperbolic tangent function. The adaptation of models to the environmental conditions was confirmed by reducing the penalty by 30.47% by the median or by 23.59% by the extreme 5% of the data, for the worst configuration. The obtained results can be used in game development, to create the “brain” of NPCs. This will allow to get rid of the exponential complexity of development, inherent in current methodologies due to the proposed approach. Using reinforcement learning technologies, it is possible to make a more variable and detailed virtual context, with lower computational resources consumption, avoiding ludonarrative dissonance.
The object of research is the spatial variability of soil salinity in the Senegal River Basin (Mauritania). The problem to be solved is to identify the most reliable spatial interpolation technique for soil salinity mapping and support the appropriate soil salinity management. This problem is especially important as soil salinization is a serious environmental and socio-economic problem that severely constrains agricultural productivity, mainly in arid and semi-arid areas. The case research was carried out on two cultivated agricultural plots and one uncultivated plot having similar soil characteristics. The four spatial interpolation methods used and compared are: global polynomial interpolation (GPI), inverse distance weighting (IDW), ordinary kriging (OK) and radial basis functions (RBF). The predictive performance of the models was evaluated using mean squared error (MSE), root mean squared error (RMSE), mean absolute percentage error (MAPE) and coefficient of determination (R2). Results showed statistically significant differences in the properties of the datasets, especially within the central tendency and distribution. Soil electrical conductivity (EC) varied substantially among the research areas, with mean values increasing from 1.35 to 5.58 dS × m–1 and spatial ranges extending up to 189 m. Across all three zones, IDW consistently provided the best interpolation performance, achieving coefficients of determination (R2) close to 0.99 and outperforming the other tested methods. This result implies that the proper mapping of soil salinity requires the selection of an interpolation method compatible with the statistical characteristics of the field data. They also stressed the need for permanent monitoring of salinity to guarantee the sustainable agricultural productivity and efficient land management in the irrigated coastal areas of the Senegal River Valley.
