Nº 3 (2025)

Background. The most important method of scientific research is classification. The task of classifying objects with a heterogeneous space of information features, which currently has no unified solution, is particularly difficult. It is important to take into account the dependence of classification results on chosen point of view on subject area – a criterion that determines composition and typology of significant features of objects under consideration. The purpose of the work is the mathematical and algorithmic formalization of the process of developing classification criteria, as well as development of a unified format for presenting information features of an object regardless of their type, composition and completeness. This will reduce classification task to a single automated process - well-managed, visual, amenable to machine learning and invariant to specifics of subject area. Materials and methods. The work reveals the main problems of traditional approaches to solving the classification problem, in context of dynamic points of view on subject area. Existing works devoted to formalization of classification criteria and problems of choosing significant information features of objects are considered. The methodological basis of the work is set theory, machine learning methods, neural networks and raster computer graphics. Results. As a result, a concept and formalized set-theoretic representation of the integral classification criterion are proposed. The criterion allows algorithmization of choice of view’s points on the studied subject area. Classification criteria determine composition of significant features of the object. For its unified representation, a single format in the form of a raster graphic image - a graphic-chromatic map, as well as an algorithm for creation of such maps have been developed. Conclusions. The proposed approach significantly simplifies the procedure of distributing heterogeneous objects into classes by reducing it to a well-studied and proven task of machine learning - classification of raster graphic images. Formalization of the integral classification criterion allows quickly and conveniently set different points of view on the subject area, representing them in the form of data tuples. The algorithm for forming grapho-chromatic maps provides obtaining raster images of a single size and format for feeding to the input of a pretrained neural network classifier. At the same time, when changing the point of view on the subject area (i.e. when changing the number and composition of significant features of objects), there is no need to change the structure of the classifier or retrain it.

Background. Based on the conducted study of developments in the field of functional architecture of cluster computing systems, the relevance of work in this direction is shown, which is due to the increasing demand for affordable high-performance computing, the growing implementation of artificial intelligence and machine learning methods. The purpose of the work is to develop and experimentally test at the level of simulation models a methodology and technology for implementing the functional architecture of cluster-type computing systems based on high-speed switches. Materials and methods. A logicalprobabilistic approach is used to create models of the functional architecture of cluster computing systems determined by applications and middleware, which allows accelerating the creation of simulation models for a number of important modes of cluster use and making the transition to logical-algebraic formalized specifications for software applications. Results. Simulation logical-probabilistic and logical-algebraic models for a number of important variants of using a computing cluster were constructed, the necessary statistical experiments with these models were carried out, which provided justification for implementing the corresponding models of the middleware software. Conclusions. A methodology for developing the functional architecture of a cluster-type computing system, defined by specifications in the form of logical-probabilistic and related logical-algebraic models, is proposed, which can accelerate the preparation of the cluster for operation in an organization.

Background. Increasing the adequacy of mathematical models of the impact of malicious software (MS), used to substantiate the requirements for the characteristics and directions for improving the anti-virus mechanisms of computer systems (CS), can be achieved by using the methodological apparatus of functional modeling. The purpose of the studyis to substantiate a methodological apparatus for formalizing the threats of MS in order to construct adequate mathematical models for assessing the temporal characteristics of such threats. Materials and methods. The solution to the problem of constructing functional models in computer systems is based on the methodology of functional modeling, systems analysis and graph theory. Rusults. A procedure for constructing mathematical models of the temporal characteristics of threats of MS has been substantiated and implemented, including: a stage of detailing the objective function “Impact of MS”, a stage of establishing the order of execution of its functional components at each level of decomposition, a stage of forming the space of MS features, a stage of presenting the model in the form of graphs and a stage of forming analytical expressions for assessing the temporal characteristics of the threat. Conclusions. The adequacy of mathematical models of the temporal characteristics of threats of such impact, achieved through the functional decomposition of the objective function “Impact of MS”, allows for the validity of requirements for the directions of improvement of anti-virus mechanisms of the CS.

Background. The object of the research is a remote energy supply complex (RESC) for spatially distributed groups of aerial objects. The subject of the research is the operational parameters of RESC and their impact on the efficiency of energy supply. The purpose of this study is to develop a mathematical model for determining optimal operational parameters of RESC and to provide a comparative assessment of the effectiveness of the proposed model in relation to existing approaches to energy supply management for spatially distributed groups of aerial objects. Materials and methods. The research was conducted using the gradient descent method to synthesize a nonlinear approximating function describing the dependence of the number of charged aerial objects on key system parameters. The adequacy of the model was verified using Pearson’s criterion. Results. A mathematical model has been developed that takes into account key system parameters: the distance of the RESC from the serviced objects, the width of the energy supply zone, and the angular magnitude of the energy supply space in the horizontal plane. The results of numerical modeling demonstrate an increase in energy supply efficiency by an average of 16 % (with peak values up to 35%) when using the developed model. Optimal parameters for the spatial placement of complexes have been established, with preferred positioning in the corner zones of the controlled territory. The average approximation error was 3.162 %. Conclusions. The developed mathematical model allows determining the optimal operating conditions of the RESC and significantly improving the efficiency of energy supply to aerial objects compared to traditional operator control. The proposed approach provides the ability to predict the effectiveness of energy supply, optimize the parameters of the location and settings of the RESC, and plan its application in various operating conditions.

Background. The object of the study is an optoelectronic system based on infrared (IR) sensors. The subject of the study is the calibration methodology of this system. The purpose of the work is to develop and present an adaptive calibration methodology that takes into account the influence of external factors (such as illumination and temperature) to improve the accuracy and reliability of measurements. Materials and methods. The research was carried out using mathematical modeling, in particular, the power approximation method and the least squares method. The system was modeled in MATLAB/Simulink and experimental studies were carried out in laboratory and field conditions. Data processing and system control were carried out using the Tiva C Series LaunchPad microcontroller. Results. The calibration method is presented, based on mathematical models describing the dependence of the output voltage of the IR sensors on the distance to the object. The analysis of the influence of external conditions is carried out. The correction factors are introduced and calculated, allowing the dynamic adaptation of the measuring system to the changing operating conditions. Conclusions. Analysis of the simulation results and experimental data showed that the proposed approach to adaptive calibration can significantly improve the accuracy and reliability of measurements. The developed methodology can be applied to improve the operation of autonomous navigation systems, medical sensor devices and industrial measuring systems.

Background. In this paper, the authors numerically evaluate the effectiveness of the previously stated technical solution to increase the rigidity of one of the basic units of a small-size milling machine - the gantry. Materials and methods. The investigated portal has dimensions 90x300x450mm and is a molded composite structure with studs in its columns as reinforcing elements. The studs embedded in the composite matrix are in a state of elastic axial tension. The calculations were carried out using finite element modeling method. To investigate the effect of stud preload on the stiffness of the structure, a series of static calculations have been carried out, where the preload value is varied. Results. The simulation results show that the use of preloaded elastically preloaded stud reinforcement results in stiffness gains ranging from tens of percent to 4,7 times, in the longitudinal direction (X axis of the machine) and up to 47% increase in the damping capacity of the structure. It is demonstrated that the increase in rigidity is nonlinearly related to the generated axial force of elastic tension of the studs and can vary in a wide range, which allows approaching and even surpassing the rigidity index of more massive and several times more metal-intensive portal structures of small-sized machines. The obtained increases in rigidity are explained by the authors as a consequence of creating a stressed connection of the portal with the bed, which provides an increase in the friction force and the force of pressing the portal to the base of the machine. The dependence of the portal rigidity on the tension of the studs is revealed and graphically shown, and the function of the dependence of deformations on the tension in analytical form is obtained. Conclusions. The proposed technical solution is promising for the small-sized machine tool industry and allows creating machine units with increased rigidity, damping and reduced metal consumption.

Background. Aluminum alloys have long proven their effectiveness as armor materials and have been widely adopted. Although they have not completely replaced traditional steel armor, in certain cases they have become an excellent alternative due to their unique properties, such as high strength-to-weight ratio, corrosion resistance, and rigidity. Currently, the most effective solutions are complex heterogeneous structures based on aluminum. However, there are several significant drawbacks associated with their production methods. These include differences in the physicochemical and mechanical properties of the base materials, interlayer corrosion, delamination, low adhesion coefficients, and others. In this context, explosive welding is considered the most promising technology for producing such materials. The purpose of this work is to evaluate the complex of physicomechanical properties of a new heterogeneous armor material based on aluminum and titanium, produced by explosive welding. Materials and methods. The primary approach to achieving the stated goal involves conducting a comprehensive study of the composite's properties: assessing the macrostructure of the material, evaluating the quantitative chemical composition of the metallic base of the composite and its heat-affected zone, and assessing bullet resistance. These studies were carried out in accordance with GOST standards using calibrated equipment by certified personnel. Results. Evaluation of the composite material's condition after explosive welding through visual and dimensional inspection, as well as analysis of its macrostructure, allowed for the selection of an optimal welding regime that ensures the highest quality of the composite without areas of delamination or incomplete bonding. Assessment of the macrostructure and quantitative chemical composition of the metallic base of the composite and its heat-affected zone provided insights into the properties and growth patterns of intermetallic phases in the weld zone. Testing the bullet resistance of the obtained composite demonstrated its compliance with the Br3 protective structure class, which, at a given thickness, is significantly higher than that of monolithic armor. Conclusions. The use of a new type of heterogeneous armor material based on aluminum and titanium, produced by explosive welding, has improved armor survivability through a novel reinforcement scheme that localizes the development of brittle cracks in the composite structure under ballistic impact. The armor material developed in this study also allows for a 20-25% reduction in the weight of armored vehicles while maintaining the required level of bullet resistance compared to monolithic aluminum armor.