This paper considers the development and training of a neural network model for the identification of the species and quantitative composition of pollen, which will subsequently be used to determine the botanical origin of honey and quantitative calculations of pollen grains contained in a certain mass of honey. The main purpose of the study is to create a model that can effectively distinguish the morphology of pollen grains present in honey, determine their quantitative and qualitative composition, which will improve product quality control, as well as identify its botanical and geographical origin. For this purpose, similar works on pollen classification were analyzed. Pwtorch was chosen as the framework for creating a neural network model, which provides the possibility of detailed configuration of the model. The result of the work is a trained model capable of classifying pollen grains.

Keywords: neural networks, classification task, pollen classification, convolutional neural networks, PyTorch

A numerical study of heat transfer and hydrodynamics in plate heat exchangers with corrugated fins was carried out. The effect of corrugation pitch on heat flow and pressure drop was studied. The study was carried out using computational fluid dynamics in the ANSYS software package (v. 19.2). The results of the study showed that increasing the corrugation pitch from n = 5 to n = 9 leads to an increase in heat flow of 10.83% and a pressure drop of 28.30%.

Keywords: plate-fin heat exchanger, corrugated fins, heat transfer, hydrodynamics, numerical study, calculation, heat flow, pressure drop, energy efficiency, cooling system

A semi-Markov stationary model of operation of measuring instruments installed on complex technical systems is developed, which allows modelling the process of metrological maintenance of measuring instruments at combined verification and operational automated condition control. The results of modelling are presented.

Keywords: semi-Markov model, measuring instrument, verification, operational automated state control

The use of small drones to detect urban targets such as vehicles at low altitude in cities has gradually become a mainstream means. In view of the existing problems of low detection accuracy of single-mode detection network caused by visible light detection, inability to work at night and blurred edge of infrared detection targets in actual scenes, In this paper, a multi-modal UAV detection algorithm based on image fusion and deep learning network is proposed. Firstly, based on DUT-VTUAV visible-infrared registration data set and TIF image fusion algorithm, multi-mode fusion data set is constructed. Secondly, by comparing the detection accuracy, speed and number of parameters of the existing YOLO series network, the lightweight network YOLO v5n which is most suitable for the mobile deployment of UAVS is selected. Finally, a multi-modal fusion detection algorithm is formed by combining image fusion algorithm and target detection model. Comparative experiments on vehicle data sets show that compared with single-mode detection, the detection accuracy of the proposed algorithm is effectively improved, with mAP up to 99.6%, and a set of visible-infrared image fusion detection can be completed within 0.3s, which has high real-time performance.

Keywords: target detection, YOLO, multimodal fusion, data fusion, TIF algorithm

The paper presents an ordinary and dynamic piecewise linear Leontief model with nonlinear predictors for the gross domestic product of Russia. The following variables are used as independent variables: the number of able-bodied population, retail trade turnover, and capital investment. In the right part of the model, along with the current value of investment volume, its lagged values with a lag of one and two years are included. The average relative error of approximation of each model and the values of the vector of triggers of independent variables were calculated.

Keywords: Leontief model, nonlinear predictors, gross domestic product, least modulus method, average relative approximation error, vector of triggers, linear-Boolean programming problem

Using numerical modeling, a study of heat transfer and hydrodynamics in plate heat exchangers with corrugated fins was carried out, while the height of the corrugation profile varied from 2 to 4 mm. The influence of profile height on heat flow and pressure drop was studied. It was revealed that an increase in the profile height leads to an increase in heat flow up to 34.05% and pressure drop up to 54.54%.

Keywords: corrugated heat exchanger, cooling system, microelectronics, profile height, heat flow, pressure drop, heat transfer, hydrodynamics, calculation, numerical modeling

A programming method based on the ESC pattern is presented. The relevance of the work is due to the fact that the object-oriented approach is one of the most popular and sought-after ways to develop an information product due to a huge regularly updated selection of various methods, templates and ways of its implementation. The most significant of them is Entity System Component (ESC). This method of implementing OOP allows you to make the software product flexible and extensible. The ESC pattern is based on the reactive programming method and divides the entire code architecture into three components: entity, system, component. The tool package that implements the ESC pattern is the ESC DOTS package, designed for the Unity3D environment. The built-in Jobs System package provides the ability to work with multi-threaded programming in Unity. This package distributes threads created at runtime into groups of a certain type, which have a strictly limited execution time. Thus, the task of enumerating an array of several hundred elements goes into a Temp type thread, which runs for one frame in Unity, and the enumeration of a million polygons of the Unity landscape is placed in a Persistent type thread, which has an unlimited time limit.

Keywords: object-oriented programming, Unity framework, ESC pattern, multi-threaded programming, reactive programming, extensible architecture, package manager

The article presents a set-theoretic model that generalizes the concept of a monitoring system. The model is a tuple that includes a monitoring object, the infrastructure of the monitoring system, initial data and monitoring results, and a set of relationships between the components of the model. Each component of the model is detailed at 1-2 levels of detail. For some elements of the model, examples from existing monitoring systems are given. The model can be used to create new or modify existing monitoring systems.

Keywords: monitoring system, monitoring object, set-theoretic model, tuple, data processing, infrastructure, sensor, software

The practical significance associated with the development of automated software systems for calculating joints of steel structures of buildings and structures is noted. Using the IDEA Statica software package as an example, the calculation and analysis of the operation of a steel unit connecting the lower chord of a truss to a column was carried out. The order and sequence of entering the initial data for calculating the connection is shown. The calculation results present in tabular and graphical forms the equivalent stresses and plastic deformations of the main elements, as well as the results of inspections of bolted and welded connections of these elements to each other. A comparative analysis of numerical and analytical calculations was performed. Conclusions are formulated based on the results presented in the article.

Keywords: calculation model, connections, steel structures, component finite element method, design, nonlinear analysis, work analysis, stress, plastic deformation, bolted connection, welded connection

The paper presents a solution to the problem of accelerating the process of visualizing the results of numerical simulation. The volumes of such data can be very large and the development of tools to speed up the process of analyzing modeling results is an urgent task. This article proposes a solution to the problem based on the development of a set of programs that automate the process of processing large-volume scientific data of the same type to create high-quality visualizations of the results of numerical modeling. The results are presented using the example of solving problems in astrophysics, but the proposed methodology can be quite easily applied to other subject areas in which models based on the dynamics of particle systems are used. The research described in the work is devoted to solving issues related to converting data obtained from numerical modeling into a format that can be read by the ParaView softwart, which implements many methods that allow obtaining very high-quality visualization. The work also describes the process of automating batch processing of a large amount of data that has the same structure, presents the results of an analysis of the acceleration of the visualization process when using the NVIDIA IndeX plug-in, and considers the possibility of improving the quality of visualization results when applying Delaunay triangulation to the original data.

Keywords: data visualization, Delaunay triangulation, rendering acceleration, ParaView, NVIDIA IndeX, VTK

Currently, there is an increase in the number of scientific papers on models, methods and software and hardware for image processing and analysis. This is due to the widespread introduction of computer vision technologies into information processing and control systems. At the same time, approaches that provide fast image processing in real time using limited computing resources are relevant. Such approaches are usually based on low-level image filtering algorithms. One of the tasks to be solved in computer vision-based systems is the localization of round objects. These objects have the property of radial symmetry. Therefore, the approach based on the Fast Radial Symmetry Transform, which is considered in this paper, is effective for solving this problem. The paper describes the basic steps of the basic transformation, provides a procedure for determining the centers of radially symmetric areas for localization of round objects in images, and discusses examples of its application.

Keywords: computer vision, image processing, image analysis, localization of objects, methods of localization of round objects, fast radial symmetry transf, detecion of the centers of radially symmetric areas

The article gives an insight into the traditional passive cooling techniques utilized in Mediterranean and hot climate regions, showcasing the effectiveness of natural ventilation and central patios in improving indoor thermal comfort and reducing energy consumption. The integration of these traditional design elements with modern technologies is crucial for creating environmentally friendly and culturally significant architectural solutions for the future. The discussion then transitions to the importance of energy efficiency and environmental sustainability in modern building design, highlighting the use of Ansys Fluent software for modeling natural ventilation in a residential apartment in Yemen. The article emphasizes the need for engineering and architectural solutions to achieve high performance in energy efficiency and environmental safety in contemporary construction practices.

Keywords: natural ventilation, Yemeni houses, Spalart-Allmaras model, civil engineering, Ansys Fluent

The purpose of this article is to analyze the atmospheric boundary layer (ABL) models used in computational fluid dynamics packages, which take into account various conditions of atmospheric stratification. The hypothesis of the study is that in this case, the boundary conditions and model parameters should ensure the a horizontally homogeneous flow in the empty computational domain. An overview of works devoted to this topic is given, as well as an example of calculation results obtained using one of the models. The review shows that the considered models used in computational fluid dynamics packages make it possible to include effects associated with atmospheric stratification and obtain horizontally homogeneous vertical profiles of ABL characteristics. It is also possible to identify issues raised by the authors of works in this area, such as modeling of a stable boundary layer, modeling cases of strong convection and stability.

Keywords: atmospheric boundary layer, computational fluid dynamics, CFD modeling, vertical profiles of meteorological parameters, boundary conditions, atmospheric stratification, k-ε-model

The article presents a model for the formation of a thermodiffusion layer on steels. Thermodiffusion coatings on 45 steel and HVG steel were obtained by a chemical-thermal method in a powder mixture. Studies of layered X-ray diffraction phase and microrentgenospectral analysis of the studied coatings have been carried out. on steel. The obtained results of X-ray diffraction and microrentgenospectral analysis of diffusion coatings confirm the correctness of the model of formation of the diffusion layer and the theoretical prerequisites outlined. The results of the study can be used in the preparation of mixtures for thermal diffusion saturation.

Keywords: diffusion layer, diffusion chrome plating, diffusion coating

This article presents the results of a study of the filtration flow in a porous material based on a thrice periodic minimum surface (TPMP) Neovius. The Brinkman model was used to determine the profile of the fluid flow velocity in the channel containing the TPMP insert. Based on computational experiments in ANSYS Fluent, the dependence of the permeability of the medium on its porosity is determined. The results of the study can be used to describe filtration processes in the oil and gas industry, hydrogeology and other fields related to filtration flows in rocks and other media.

Keywords: permeability, filtration, porous medium, minimum surface, TPMP, ANSYS, Brinkman model, numerical modeling, boundary value problem, velocity profile

Wavelets are widely used in various fields of science and technology for processing 1D signals and multidimensional images. However, technical information processing devices are developing more slowly than the amount of digital data is growing. Computational latency is the most important characteristic of such devices. This paper proposes an implementation of the Winograd method with a convolution step 2 to reduce the latency of wavelet image processing. The proposed scheme for implementing calculations has reduced the asymptotic computational complexity of wavelet processing of 2D images to 53% compared to the direct implemettaion method. A theoretical assessment of the computing device characteristics showed a reduction in latency of up to 67%. A promising direction for further research is the hardware implementation of the proposed approach on modern microelectronic devices.

Keywords: image processing, Winograd method, digital filtering, computational delay, wavelet transform, convolution with step

The paper considers a mathematical model of a Darye wind turbine with three blades. The operating mode of the fast rotation of the wind turbine has been obtained. It is called autorotation mode. An assessment of the speed of the wind turbine in this mode was obtained, an assessment of the power and geometric dimensions of the structure under consideration was carried out. Programs have been written in the matlab system and numerical calculation results have been obtained. The efficiency of Darye wind turbines with a different number of blades is compared and it is shown that in practice it is best to use three blades.

Keywords: body, planning mode, stability, geometric dimensions

Modeling the behavior of cylindrical shell structures weakened by holes when applying a load evenly distributed over the surface, taking into account geometric nonlinearity in the finite element software package ANSYS Mechanical APDL 2022R. Models of cylindrical shell structures with various hole arrangements have been created, and an analysis of the behavior of structures with an increase in the magnitude of the acting external uniformly distributed load has been carried out. Material type: isotropic, steel. Load-deflection graphs are obtained. The values of the critical load at which the structure loses stability are found. The values of internal stresses arising in structures under the influence of a critical load are found. The values of the displacement of the nodes are found: central and quarter under the influence of a critical load.

Keywords: Shell structures, stress-strain state, loss of stability, finite element method, deflection, supercritical behavior, geometric nonlinearity, step-by-step application of load, load–deflection dependence graph, isotropic material

The activities of modern enterprises and organizations are becoming increasingly complex, which complicates decision-making tasks. In this regard, the issues related to the integration of innovative information technologies in the field of decision-making are being updated. One of these technologies is artificial intelligence, the main advantages of which are the ability to solve time-consuming tasks and perform analysis based on big data in real time. The purpose of the work is to analyze the methods of applying artificial intelligence technologies in the theory of decision-making. The paper considers the fundamental factors associated with the work of artificial intelligence, prospects and current trends in its development in solving decision-making problems. The scientific value of the article consists in an attempt to systematize knowledge regarding the use of intelligent technologies in decision-making theory. The materials of the article can be useful for modern software developers and the management of organizations.

Keywords: artificial intelligence, algorithm, data analysis, decision-making, model, information technology

In this article we present a novel algorithm for detecting moving objects using a stationary camera, which is based on statistical background modeling using edge segments. Traditional algorithms that rely on pixel intensity struggle in dynamic environments due to their inability to handle sudden changes in lighting conditions. These algorithms also tend to produce ghosting artifacts when a sudden change occurs in the scene. To address this issue, edge-based features that are robust to intensity variations and noise have been introduced. However, existing methods that rely on individual edge pixels suffer from scattered edge pixels and cannot effectively utilize shape information. Additionally, traditional segment-based methods struggle with variations in edge shape and may miss moving edges that are close to the background edges. In contrast to conventional approaches, our proposed method constructs the background model using regular training frames that may include moving objects. Furthermore, it avoids the generation of ghosting artifacts. Additionally, our method employs an automatic adaptive threshold for each background edge distribution to facilitate matching. This enhances the robustness of our approach to changes in illumination, camera movement, and background motion. Experimental results demonstrate that our method outperforms other techniques and efficiently detects moving edges despite the aforementioned challenges.

Keywords: motion detection, edges, canny edge detector, gaussian of color, gaussian of gradient magnitude, normal distribution, adaptive thresholds, statistical map

Analyses for the current publishes show that the problem of forecast water overflowing is actual and often causing a lot health threaten and other dangerouses. This article offers computing, analysis and development the regression model of the level of Ia river. The final model correspont the real data with proper level. The final calculation means that this model could be used for real forecast for defend the people from water's overflow.

Keywords: model, simulation, river, water level, flood, emergency, forecast, statistics, monitoring, analysis, iya river, Irkutsk region

A combinatorial problem - the problem of finding the set of all formal concepts of formal concepts is considered. The computational complexity of the problem is that the number of formal concepts depends exponentially on the size of the initial formal context. in the article to solve this problem, an algorithm for fragmentation and defragmentation of the formal context is given, based on the method of decomposing the formal context into fragments. The essence of the method is that the original formal context is divided into various fragments. The fragments have different sizes and a non-empty intersection. Each fragment is subsequently considered as a formal context and can again be subject to decomposition. As a result, a finite set of fragments is formed. Then formal concepts are found in each fragment and combined to form the desired set of all formal concepts of the formal context. The method is “non-distorting”: when dividing the context into fragments, new formal concepts are not formed and the sought-for concepts are not lost. The results of computational experiments are presented, showing the effectiveness of the developed algorithm.

Keywords: formal concepts analisys, fragmentation algorithm, formal context, object-attribute table, combinatorial problem, the problem of finding the set of all formal concepts

In this study, an analysis of the time series was conducted using a class of shift functions for arithmetic and geometric progressions, along with their synchronization using logarithmic decrement. The closing prices of IBM company stocks were taken as the examined data for each trading day. The shift functions of geometric and arithmetic progressions revealed almost-proportions and almost-periods in the examined data. These detected patterns emphasize the importance of applying shift functions in the analysis of time series, allowing the extraction of internal patterns and periodic fluctuations that might go unnoticed with standard analysis methods. Computing the minima and corresponding values of the geometric progression enabled the identification of almost-periods in the data. These results not only confirmed visual observations but also enhanced our understanding of the internal patterns of the time series. The findings underscore the effectiveness of applying methods for analyzing time series based on almost-proportions and metric techniques. These approaches play a crucial role in uncovering hidden patterns and subtle periodicities in data, providing a fundamental foundation for more accurate analysis and successful forecasting.

Keywords: nearly-proportionalities, synchronization of geometric progression, empirical data, geometric progression, shift functions

In connection with the revival of interest in wooden housing construction in Russia, the creation of specialized software systems for the calculation and design of buildings and structures made of wood is becoming relevant. One of the most popular systems in the world is light-frame housing construction. As one of the possible tools for automated design of light-frame wooden buildings, this article presents the FrameCAD program, developed at the Department of Metal, Wood and Plastic Structures of the Don State Technical University. Some of its capabilities are presented, allowing you to design wall fences of light-frame buildings, beamed wooden floors, perform calculations of individual frame elements taking into account the requirements of current regulatory documents and automatically generate albums of drawings of wooden structures in the domestic NormCAD program, as well as specifications for them factory made. FrameCAD has been tested in the design of a number of objects. The software product is developed by engineers who have many years of experience in the actual design and construction of light-frame buildings. It continues to be improved and supplemented, and in the future it may become a replacement for similar foreign complexes.

Keywords: wooden structures, software package, wall panels, beam overlap, automated calculation

The article considers the issue of studying the deviation of geometric parameters of a curved surface and a facet surface approximating it. The approximating surface was constructed by triangulation. To obtain adequate results when approximating a surface, it is necessary that the geometric parameters of the approximated surface differ minimally from the parameters of the faceted surface. One of the most important geometric parameters of a surface is its normal at a given point. The deviation of the normals of the given and approximating surfaces significantly affects the calculation error.

Keywords: curved surface, faceted surface, normal to the surface, linear frame, tinted image, plane compartment