Purpose of research. Overhead communication lines (OCL) are an important element of the communication infrastructure, but their technical condition requires regular monitoring and inspection. Traditional inspection methods, including visual inspection by specialists, do not always allow for the efficient collection and recording of all necessary data. In order to improve the quality of OCL inspection, a method was developed for determining the tilt of a communication line pole based on images from an unmanned aerial vehicle (UAV).

Methods. A combination of mathematical transformations and machine learning methods was used to solve the problem. Data processing included the use of camera parameters, object coordinates in the image, flight altitude, and UAV coordinates. Based on these data, an algorithm was developed for detecting key support points and calculating the tilt angle of the poles.

Results. As a result of the experiments conducted based on the data obtained from the UAV, the accuracy of detecting key support points according to the mAP50 metric was 0.71. Within the correctly predicted support, the accuracy of detecting its top and base was 0.88 according to the F1-score metric. To determine the tilt of the VLS pillars, a formula was derived that made it possible to calculate the maximum tilt of the pillar is 24.5°, and the minimum is 0.6°. The average tilt angle of the pillars for the entire set of images is approximately 6.1°.

Conclusion. The developed method allows automating the technical inspection of VLS, ensuring high accuracy in determining their key parameters. The use of UAVs and machine learning reduces time and cost, and improves the quality of data collection and analysis. The use of UAVs in combination with machine learning methods can significantly reduce time and cost, improve the quality of data collection and analysis, and reduce the risk of human error.

Purpose. The study presents the key aspects of the introduction and use of artificial intelligence (AI) in the construction industry. The challenges faced by builders when implementing AI, such as the high cost of equipment and the need for retraining, are discussed. The advantages of AI are also considered, including improving design efficiency, reducing costs and improving control over construction processes. Special attention is paid to the use of AI in monitoring construction sites, risk forecasting, automation of typical operations and the use of autonomous technology. The article emphasizes that despite the existing problems, it is a powerful tool for the transformation of the construction industry, which can significantly improve the quality and speed of construction.

Methods. The research uses methods of statistical and comparative analysis, the «golden square» method in the context of digitalization of the IC, focused on managing complex situations.

Results. A classifier of threats and risks for IC in difficult situations has been developed from the point of view of the sustainable development of artificial intelligence in the construction industry.

Conclusion. Artificial intelligence in construction opens up new horizons for improving efficiency and optimizing processes. From BIM technologies to autonomous robots and drones, and is becoming an integral part of the modern construction industry. However, despite all the advantages, there are also challenges, such as the high cost of equipment and the need for retraining. Nevertheless, the potential in construction is huge, and its implementation promises significant improvements in the quality, speed and safety of construction work. The paper provides an overview of the problems of IC in modern conditions. The analysis of the application of the "golden square" method in the context of digitalization of the IC, focused on managing complex situations, is presented. A classifier of threats and risks for the UK in difficult situations from the point of view of sustainable development has been developed.

Purpose of research. For the geometric parameters of the device obtained as a result of studies, which improve the conditions for the formation of the suction flow. and the proposed design solutions of an exhaust device for local ventilation systems to remove harmful emissions during plasma metalworking, which make it possible to improve working conditions and perform industrial tests of the device sample.

Methods. Numerical modeling was performed to obtain a spatial representation of the current lines and velocity fields. Physical modeling methods were used to obtain dependencies that formulate methods for calculating the aerodynamic and ecological-energy parameters of the resulting flow during the interaction of the flow, the resulting hazards with the suction flow and the radial limiting ramjet.

Results. Confirmation of theoretical studies of the design of a local exhaust device with a limiter diffuser designed to form a limiting radial jet and experimental studies on tests of an industrial sample of a local exhaust device of the proposed design have been obtained.

Conclusion. The recommended local coaxial exhaust device of the proposed design can significantly improve the microclimate parameters at stationary plasma cutting sites for a wide range of machines for automated processing of various metals during plasma cutting of metals. Based on experimental and theoretical studies of a coaxial exhaust device, which implements the principle of removing hazards from the plane below the workpiece being processed, limited by a radially directed distributing flow, axial velocity values for various air flow rates with an optimal ratio of suction and limiting flows β for the patented design of a local exhaust device were obtained for the first time.

Purpose of research. The problem statement is inspired, among other things, by the fact of the evolution the existing World Wide Web into the Semantic Web. The aim of this work is to develop a multi-agent environment with agents of different types: cognitive and deductive reactive. It is shown that the knowledge and beliefs storage of cognitive BDI agents (BDI is an abbreviation for Belief–Desire–Intention) in the form of facts of an extensional database can be used to obtain new knowledge and beliefs by deductive inference in the environment of an intensional database of agents possessing deductive presumptions.

Methods. The operation of an intelligent system is described by an incompletely defined conceptual graph and a system of production rules suitable for representing deductive presumptions of agents in a high-level declarativeimperative programming language. Reactive agents implement their rational behaviors based on deductive abilities, which are understood as the ability to build correct conclusions.

Results. The implementation of an intelligent agent-based system is proposed. At the conceptual level, the architecture of the intelligent agent-based system is proposed to be represented by three sublevels. The cognition of agents working at the first and second sublevels is associated with modalities. Based on the formulated interpretations at the third sublevel, reactive agents implement their rational behavior based on deductive abilities, which imply the ability to build correct conclusions.

Conclusion. It is shown that the software implementation of cognitive presumptions can be performed using conventional imperative programming languages and, possibly, data manipulation languages. The implementation of the functions of deductive reactive agents when performing operations with conceptual graphs is demonstrated using the example of filling in missing relations in a specific subject area.

Purpose of research. Analysis of system relationships and identification of patterns between the characteristics of a subscriber of an energy supply company and the amount of his receivables based on data analysis models and tools of the Loginom low-code platform.

Methods. A classification of areas for the analysis of accounts receivable in power supply companies is given. It is shown that the study of accounts receivable situations in the context of the state of the organization and an individual debtor is possible on the basis of the data of the accounting system. For a detailed analysis of the register of debtors, it is necessary to consolidate data on subscribers, payments, as well as on objects of civil law. The choice of methods for analyzing the register of debtors of an energy supply company is substantiated. The process of creating a portrait of debtors is described. Problems of the initial data are revealed. The requirements for technologies and tools for the automation of analytical processes are formulated.

Results. A data processing scenario based on the low-code Loginom platform has been developed, which includes 6 logical blocks (notes) and 2 submodels. The Outlier Editing submodel is designed to improve the quality of the source data. The "Debtor Portrait" submodel based on correlation analysis and data clustering helps to obtain information for compiling a debtor portrait. With the help of cluster silhouettes, it is proposed to assess the quality of the constructed model "portrait of the debtor". Using the ABC analysis handler, subscribers with low payment discipline were ranked into categories. Analytical processes for making management decisions in relation to debtors of an energy supply company have been built.

Conclusion. An algorithm for the analytical process of working with debtors of an energy supply company has been developed. The information and analytical system "Debtor Analysis" based on the Loginom low-code platform was implemented. Tools for making management decisions based on the results of the analysis of the register of debtors have been implemented.

Purpose of research. Increasing the level of security of information processed in information systems based on microservice architecture; by creating an effective protection system designed on the basis of knowledge obtained as a result of creating an attack surface model.

Methods. During the analysis, types of information systems (IS) were considered, among them complex IS created on the basis of microservice architecture were highlighted. Russian and foreign technologies, software allowing to automate the process of information processing were considered. A set-theoretic model of constructing an attack surface for information systems built on the basis of microservice architecture was proposed.

Results. An original approach to the description of the attack vector and surface is proposed, including a list of frequently encountered vulnerabilities, methods and tools for implementing an attack, as well as a list of possible objects of influence. A set-theoretic model for constructing an attack surface for information systems built on the basis of a microservice architecture is developed.

Results. Conducting research and developing an attack surface model for complex information systems built on a microservice architecture will improve the level of knowledge in the field of information security (IS) and ensure the security of processed data by building an effective information security system that takes into account current threats and methods of influencing the information system.

Purpose of research. The purpose of the research is to develop a generalized method for estimating the geometric parameters of an object from photographs in the absence of information about the a priori parameters of the object, as well as with the possibility of using digital cameras with various optical systems to produce images.

Methods. To achieve this goal, a mathematical model of an optical system consisting of two digital cameras was developed. Then, by reducing this mathematical model to a single system of equations, a method was developed for calculating the coordinates of an object in an image, as well as the distance to it. Then the experimental verification of the obtained method was carried out.

Results. The result of the work is a generalized method for estimating the geometric parameters of an object in the absence of information about the a priori parameters of the object using two images from digital cameras with different optical systems. An experimental verification of the accuracy of the developed method was performed when calculating the distance to an object using standard cameras with different lenses, which showed that this method allows estimating the geometric parameters of objects with an accuracy of 90% or higher, depending on the accuracy of measuring the technical parameters of the cameras. The sensitivity of the developed method to the values of the technical parameters of the digital cameras used was also evaluated.

Conclusion. Experimental tests have shown that the developed method allows us to accurately estimate the distance to objects using images from two digital cameras, however, it requires high accuracy in measuring the technical parameters of the cameras used. As further ways of developing this method, it is possible to note the possibility of increasing the accuracy of measurements by adding nonlinear distortions (distortions) to the developed mathematical model.

Purpose of research. The article investigates a method proposed by the authors for reducing motion blur in images of objects moving at high speed on a conveyor belt, using a moving camera whose speed is synchronized with the movement of the conveyor belt. Application of this method improves the quality of the obtained images and, consequently, the efficiency of automated quality control and object identification systems on the conveyor. To assess the degree of image blurring, a metric based on the analysis of the frequency spectrum of the obtained image was used.

Methods. A method for reducing motion blur in object images has been developed, based on the use of an automatic synchronization system that equalizes the speeds of the camera and the object on the conveyor at the moment the image is captured. To ensure reciprocating motion of the camera, it is mounted on sliders of a self-balancing double crank-slider mechanism (CSM). The article presents the structure of the automatic synchronization system and provides a description of its operating algorithm. A condition for synchronizing the movement of the camera and the conveyor was derived. To test the machine vision system with a moving camera mounted on the CSM slider, a test prototype of the mechanism providing reciprocating motion of the camera was constructed.

Results. Using the test prototype of the mechanism, a comparison between a machine vision system with a static camera and one with a moving camera at various conveyor belt speeds was made. The obtained results show that the moving camera significantly reduces the effect of blurring, especially at high speeds.

Conclusion. The proposed method allows for a significant reduction or complete elimination of motion blur, leading to a substantial improvement in the quality of the obtained images and enhancing the overall efficiency of the machine vision system. It is noted that the system requires precise synchronization accuracy between the camera and the conveyor, and its implementation may involve certain technical challenges. Nevertheless, the obtained results open up broad prospects for further development and application of the method in various fields requiring high-precision and reliable visual information about moving objects.

Purpose: improving the speed and accuracy of temperature measurement by a resistive sensor (RTD) with remote two-wire connection in distributed monitoring systems. Development and implementation of a temperature measurement method based on the processing of the integration results of the initial phase of the transient discharge process of a capacitor shunting a resistance thermometer, evaluation of the parameters of the integration time determination model and testing of the method on an experimental bench. Determination of errors in measuring RTD resistance by integrating the initial phase of the transition process using a linear model for determining the integration time and evaluating the effectiveness of the proposed solution in comparison with alternative methods.

Methods. The mathematical description of the method is based on the theory of electrical circuits. The effectiveness of the method was evaluated based on the results of experimental studies. When developing a linear model for determining the integration time, a linear regression model was built, and relative errors were calculated based on the average results of multiple measurements.

Results. A method for determining the resistances of resistive temperature sensors based on the processing of the integration results of the initial phase of the transient capacitor discharge process on a two-wire connection in distributed monitoring systems is proposed and investigated.

A mathematical description of the method is given, on the basis of which an algorithm for calculating the resistance of the RTD has been developed, eliminating the influence of the resistance of the connecting wires on the measurement results. The developed algorithm is based on the integration of the transient process of capacitor discharge (accumulation and summation of samples) over a limited time interval, while preserving the results in the middle (t1) and the end of the interval (t2) and calculating the resistance of the RTD based on the obtained parameters.

The parameters of the integration time adjustment model are determined, and measurement errors are estimated. The method was tested using an experimental stand based on an ATmega328 microcontroller and a P4831 resistance store with an accuracy class of 0.02.

Discussion. The results of the research and testing of the RD temperature measurement method presented in the paper demonstrate its effectiveness in reducing measurement errors caused by the influence of the resistance of connecting wires. The application of the proposed measurement methods and processing algorithms makes it possible to use two-wire sensor connections in distributed monitoring systems while maintaining measurement accuracy at the level of more complex and expensive threeand four-wire circuits, eliminating the influence of the resistance of the connecting wires. The application of the voltage integration method described in the work at the initial stage of the transient process allows not only to increase performance, but also to ensure the required level of measurement accuracy. Experimental studies have shown that the relative measurement errors of the method proposed by the authors when using the linear model for determining the integration time do not exceed 0.07% in the range of nominal resistances of 1-4 kOhm (corresponds to the temperature range measured by a platinum resistance thermometer, 0600 0C) with an artificial increase in the total resistance of the connecting wires to a value exceeding 200 ohms.

The proposed method can be applied in monitoring systems using taxiways located at a considerable distance from the measuring unit.

Purpose of research. Methodology and analysis of system reliability models, including monitoring the condition of individual components and failure assessment based on the failure rate and the probability of failure-free operation of modules. The work is aimed at creating effective methods for predicting failures and estimating system recovery time, taking into account the failure parameters of individual modules and their interaction. The work also involves modeling the monitoring of the state and functioning of the system.

Methods. Research methods include analyzing system reliability using failure rates, modeling the probability of uptime of system components, calculating the average time to failure, using probability distributions to determine system status, monitoring system status in real time, estimating system recovery time, as well as developing algorithms for detecting failures and managing system recovery, modeling methods.

Results. The results of the study show that as the system's operating time increases, the probability of uptime for each component decreases due to the accumulation of failures. The simulation showed that the reliability of the system depends on the failure rate of each module and their relationship in the system. For three different systems with different failure rates, the change in the probability of uptime (R(t)) was calculated as a function of time. The analysis showed that as the operating time increases, the probability of system failure increases, which is confirmed by a decrease in the value of R(t) as the time t increases. When modeling system recovery, it was found that the recovery time depends on the number of failed modules and the intensity of their failures. The monitoring system successfully reacts to changes in real time, identifying significant events such as module failure, and estimating system recovery time based on current component status data.

Conclusion. In the course of the research, a method for modeling system reliability based on an analysis of the failure rates of its components and the probability of uptime (R(t)) was developed and implemented. The methods considered made it possible to calculate the average time to system failure, as well as to estimate the system recovery time after failure, which is a key factor in ensuring the smooth operation of critical systems. The effectiveness of the monitoring system has been demonstrated, which, based on data on the state of components (M(t)), can quickly identify failure events and predict recovery time, which allows for prompt measures to restore system operability.