РУС
ENG
№ 2(37)'2025
Aims and Scope
Founder and Publisher
Editorial board
Licensing terms
Privacy Statement
Plagiarism policy
Publication ethics
Archiving Policy
Subscription
Instructions for authors
Reviewing proccess
Copyright Notice
License agreement
Article processing charges
All issues
Search
Contacts
Section
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
named after V. A. Antipov"
METHODS FOR IMPROVING THE ACCURACY OF OBJECT CLASSIFICATION WHEN BUILDING A VISION SYSTEM FOR EXPLORING CONFINED SPACES
UDC: 519.4
DOI: 10.24412/2413-7383-2025-2-37-4-12
Language: Russian
Abstract: The article discusses methods for improving the accuracy of object classification in vision systems operating in confined spaces. Approaches to the synthesis of neural network models based on logical neurons and sigmoidal activation functions are proposed, providing adaptivity to incomplete and noisy data. The concept of fractal structure (FFS) and cooperative neural network architecture, combining private models to improve the accuracy of solutions, has been developed. The presented methods increase the interpretability and stability of neural networks, which is especially important for intelligent navigation and analysis systems in technically difficult conditions.
Keywords: Artificial neural networks, classification of objects, fractal structure of neural networks, vision systems, logical neurons.
A. V. Nicenko, V. Ju. Shelepov
A METHOD OF ACCUSATIVE AND INSTRUMENTAL CASES DISAMBIGUATION FOR A NUMBER OF RUSSIAN NOUNS
UDC: 004.89:004.933
DOI: 10.24412/2413-7383-2025-2-37-13-20
Language: Russian
Abstract: The article proposes a method for automatic disambiguation of accusative and instrumental cases for a number of nouns using verb control data obtained from the Russian National Corpus. The results are implemented in experimental software for disambiguation.
Keywords: natural language processing, disambiguation, noun, accusative case, creative case, verb forms.
1Federal State Budgetary Institution "V.K. Gusak Institute of Emergency and Reconstructive
STUDY OF APPLICATION OF MASK R-CNN AND SEGMENT ANYTHING MODEL (SAM) FOR INSTANCE SEGMENTATION OF MESENCHYMAL STEM CELLS IN MICROPHOTOGRAPHS
UDC: 004.93
DOI: 10.24412/2413-7383-2025-2-37-21-29
Language: Russian
Abstract: The study compares the efficiency of Mask R-CNN and Segment Anything Model (SAM) for instance segmentation of mesenchymal stem cells (MSCs) in microphotographs. Results on two datasets (low and high cell density) demonstrate that Mask R-CNN, fine-tuned on target data, outperforms SAM in mAP (0.64 vs. 0.61 for low density; 0.31 vs. 0.26 for high density). Increased cell density and overlap reduce accuracy for both models. SAM’s limitations in using only bounding boxes highlight the need for domain-specific adaptation.
Keywords: instance segmentation, mesenchymal stem cells, morphological heterogeneity, cell density, automated cell culture analysis, biomedical imaging, model adaptation.
Section
MATHEMATICAL MODELING, NUMERICAL METHODS AND SOFTWARE SYSTEMS
A. I. Borovikov, O. A. Krivodubsky
MATHEMATICAL MODELING AND FLOW FORMALIZATION FOR ANALYSIS, OPTIMIZATION AND ANOMALY DETECTION IN LOGISTICS, FINANCE AND PROCUREMENT
UDC: 519.8:658.512
DOI: 10.24412/2413-7383- 2025-2-37-31-43
Language: Russian
Abstract: The article provides an overview of modern methods for formalizing material, financial, and information flows in enterprises. It examines object -oriented and process -oriented approaches, as well as discrete, graph -based, and matrix models used for analysis, optimization, and anomaly detection in logistics and financial processes, including procurement. The article discusses the integration of various models and the use of machine learning tech niques to enhance analysis efficiency and decision-making.
Keywords: mathematical modeling, flow formalization, logistics, finance, procurement, anomalies, optimization, machine learning, graph models, matrix models, discrete models.
V. S. Solod
PHYSICAL AND MATHEMATICAL MODEL FOR DEVELOPING AN EXPERT SYSTEM FOR ACCELERATED COOLING OF REINFORCEMENT PRODUCTS
UDC: 621.771.07:004.94
DOI: 10.24412/2413-7383-2025-2-37-44-58
Language: Russian
Abstract: The ultimate goal of the study is to develop an expert system for accelerated cooling of reinforcing steel in the form of an artifi cial neural network (ANN), using a new machine learning method: using fundamental knowledge to solve practical problems. A physical and mathematical model of accelerated multi -stage cooling of reinforcing steel in a rolling mill flow is used as fundamenta l knowledge. This model can be used both as a tool for creating a training data set for training ANN, and for optimizing the technology for cooling profiles from various steel grades with possible changes in rolling modes.
Keywords: hardening, reinforcement bars, thermal strengthening, physical and mathematical model, algorithm, thermal facility
Section
SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING, STATISTICS
S. S. Antsyferov, K. N. Fazilova, D. S. Muravyov
INTELLIGENT QUALITY CONTROL SYSTEMS FOR HIGH- TECH MANUFACTURING PROCESSES AND EQUIPMENT
UDC: 681.518.9; 621.384.32
DOI: 10.24412/2413-7383-2025-2-37-60-65
Language: Russian
Abstract: В статье предложена методология управления качеством технологических процессов и оборудования наукоёмкого производства. Практическая апробация данной методологии показала возможность ее использования при сопоставлении текущей ситуации с ранее просчитанными вариантами из дерева всех состояний качества.
Keywords: methodology, quality management, technological process, technological equipment, neural network, machine learning, quality control.
A. V. Zviagintseva, I. Y. Kovalev
ANALYSIS OF CLASSIFICATION MODELS FOR RECOGNIZING PRECEDENT EVENTS IN OIL AND GAS PRODUCTION PROCESSES
UDC: 004.891.3: 005.332.7
DOI: 10.24412/2413-7383-2025-2-37-66-78
Language: Russian
Abstract: The article provides a comparative testing of event classification models using a dataset of accidents at oil wells with natural oil and gas inflow as an example. The data were analyzed and processed, the hyperparameters of each model were adjusted, the models were trained, tested, and metrics were calculated.
Keywords: events, oil production, binary classification, machine learning.
K. A. Nikitenko, A. V. Zviagintseva
INTERPRETABILITY OF NEUROSEMANTIC MODELS IN THEIR APPLICATION IN APPLIED FIELDS
UDC: 004.8+004.912
DOI: 10.24412/2413-7383-2025-2-37-79-90
Language: Russian
Abstract: The article addresses the problem of interpretability in neurosemantic models used for natural language processing tasks. It discusses the main challenges related to the interpretation of high - dimensional vector representations, context -dependent features, and deep transformer archi - tectures. The paper outlines current approaches to interpretability, including both intrinsic and post - hoc methods. Special attention is given to the trade-off between accuracy and interpretability, as well as to the prospects for further development in this area.
Keywords: interpretability, neurosemantic models, applied areas, transformer, explainability, practical application.
2 JSC M. A. Kartsev Research Institute of Computing Systems
CREATION OF A MIVAR EXPERT SYSTEM FOR AUTOMATION OF FREQUENCY CONVERTER SELECTION
UDC: 004.891+007.52+004.896+681.518+65.011.56
DOI: 10.24412/2413-7383-2025-2-37-91-104
Language: Russian
Abstract: The article investigated the problem of selecting frequency converters. The relevance of the work was formulated, and the development of a mivar expert system (MES) was proposed as a solution, the task of which would be the automated selection frequency converters based on user needs. The expert system has been developed using advanced technologies of logical artificial intelligence based on the tool "Wi!Mi Razumator". The basis for decision-making will be the mivar knowledge base, for which the selection sequence of frequency converters has been described. The result of the work is an MES for automating the selection of a frequency converter.
Keywords: mivar, frequency converter, mivar expert system, knowledge base, Wi!Mi Razumator.
Section
AUTOMATION AND CONTROL OF TECHNOLOGICAL PROCESSES AND PRODUCTION
D. V. Bazhutin, V. N. Pavlysh
ESTIMATION OF RESIDUAL PAYLOAD SWING AMPLITUDE FOR AUTOMATED CRANE UNITS SUBJECTED TO INACCURATE CABLE LENGTH VALUE
UDC: 519.4
DOI: 10.24412/2413-7383-2025-2-37-106-115
Language: Russian
Abstract: The paper considers the problem of improving the payload swing elimination algorithms for gantry cranes implementing stepwise acceleration diagrams for horizontal motion electric drives. The analysis inaccurate cable length estimation effect on the accuracy of three-stage payload swing elimination algorithm for gantry crane automated motion control systems is given. The analytical dependencies of amplitude and starting phase of payload residual sway from relative cable length estimation error and electric drive acceleration parameters were derived. The correctness of obtained results was confirmed via computer simulation.
Keywords: gantry crane, anti-swing control, three-stage acceleration diagram, residual sway, computer simulation.
N. V. Vodolazskaya, E. V. Kopytova, K. V. Lavrentieva, V.D. Chernenko
INFORMATION SYSTEM FOR ACCESS CONTROL AND TRACKING OF PRODUCTION PROCESS STAGES
UDC: 658.011.56
DOI: 10.24412/2413-7383-2025-2-37-116-128
Language: Russian
Abstract: The paper proposes a comprehensive information system based on RFID technologies to increase the transparency of production processes. The system combines product movement monitoring and personnel access control. Analysis of the technological cycle was carried out, functional models and database structure were developed. The solution provides complete product traceability and labor accounting.
Keywords: information system, RFID technology, traceability of movement of components.
V. Iu. Podlesnyi, E. V. Kolesnik, I. A. Tretiakov., V. V. Danilov
OPTIMAL INTERPOLATION OF CONTINUOUS MEASUREMENT PROCESSES IN ACS AND ASRS
UDC: 65.011.56:681.5.08
DOI: 10.24412/2413-7383-2025-2-37-129-147
Language: Russian
Abstract: This paper continues the results of the study of the basics of information and measurement technologies in automated control systems and automated scientific research systems, in particular, the discrete form of representation of continuous measurement processes, namely: the optimal method of interpolation, evaluation of interpolation methods; continuous messages in the form of regular samples during least squares reconstruction; difference and delta -discrete representations; the error of discrete representation by the indicator of uniform approximation; extrapolation errors.
Keywords: ACS, ASRS, IMS, measurement messages, information characteristics, discrete representations, interpolation, extrapolation, errors, samples, errors
E. V. Skodtaev
HARDWARE ARCHITECTURE OF THE INFORMATION AND CONTROL SYSTEM OF A MOBILE PRIMARY MILK PASTEURIZATION UNIT
UDC: 65.011.56:681.5.08
DOI: 10.24412/2413-7383-2025-2-37-129-147
Language: Russian
Abstract: The development addresses the practical necessity of automating milk pasteurization processes in geographically isolated mountainous regions where centralized infrastructure is often lacking. The core engineering challenge is the formulation of a fault -tolerant hardware architecture that ensures autonomous operation with minimal dependence on human intervention. The proposed solution adopts a structured three -tier model: at the field level, sensors and actuators handle real - time physical measurements and process control; the control level utilizes a Modbus TCP/IP protocol stack to manage logic operations and facilitate communication between distributed modules; t he monitoring level integrates an HMI interface with 4G connectivity to enable remote supervision, data logging, and diagnostics. As a result, the system delivers a cohesive framework for signal acquisition, control execution, data transmission, visualizat ion, and persistent storage. The architecture demonstrates its suitability for deployment in mobile milk processing stations, particularly under conditions of logistical and infrastructural constraints.
Keywords: hardware architecture, milk pasteurization, automation, Modbus TCP/IP, remote monitoring.
3 JSC M. A. Kartsev Research Institute of Computing Systems
MES FOR SUPPORTING DECISION-MAKING OF GAS PROCESSING SHOP STAFF IN LIQUID SULFUR MINING
UDC: 004.891+007.52+004.896+681.518+65.011.56
DOI: 10.24412/2413-7383-2025-2-37-159-172
Language: Russian
Abstract: This article presents the development of a mivar expert decision support system for the personnel of a gas processing shop during liquid sulfur extraction. The system is implemented using mivar technologies of logical artificial intelligence and the KESMI platform. An analysis of technological factors affecting the efficiency of liquid sulfur extraction was conducted, key parameters determining production risks and violations were identified, and a forecasting model was developed to support operational and strategic decisions of personnel. The proposed approach allows minimiz ing risks, increasing productivity and ensuring the stability of the technological process.
Keywords: mivar, mivar networks, mivar expert systems, machine learning, technological process, production, Claus process, KESMI Wi!Mi RAZUMATOR, mivar technologies
