This article examines the practical implementation of a methodology based on Bayes' theorem in the field of technical diagnostics and residual life forecasting for industrial equipment. Emphasis is placed on the ability of this approach to support engineers' effective work under the uncertainty inherent in real-world production processes. Using a vibration monitoring system for multi-axis milling machines, which are critical for the aerospace industry in the production of high-precision aluminum aircraft components, as an example, the feasibility of quantitatively updating the probability of failures as new sensor data arrives is demonstrated. Initial signals, such as vibration, temperature, or acoustic emission levels, are transformed into probabilistic risk assessments with practical justification, providing a reliable basis for management decision-making.

 

Keywords: "Technical diagnostics, residual life forecasting, predictive maintenance, engineering systems, decision theory, aircraft engineering, repair economics, uncertainty management, probabilistic models, monitoring systems, adaptive algorithms

The article examines the main aspects of confidentiality in generative multimodal systems, describes the mechanisms and methods of ensuring confidentiality. The analysis of a method for quantifying the level of confidentiality of generative multimodal systems is carried out. STRIDE, TRIKE, OCTAVE, PASTA, and VAST are among the most advanced approaches to building security models. An analysis of all the aforementioned policies, security models, threat models, and security criteria was conducted to develop a model for privacy protection. In the process of threat modeling using the STRIDE methodology for each of the threat categories, it was found out that existing threats arise primarily due to the lack of multifactor authentication, which could protect against attempts by an attacker to obtain information and create bypass accounts in case the main one is blocked. A general approach to assessing the confidentiality of the system is proposed. Each privacy metric meets a set of criteria by which the GMS privacy is assessed.
Introduction
One of the components of the reliability of computer systems is their confidentiality. Confidentiality in generative multimodal systems (GMS) is reduced to ensuring that private or other confidential information is protected, as its unauthorized disclosure can lead to significant material losses.
Aims and objectives
The purpose of the article is to consider the issues of ensuring confidentiality in generative multimodal systems and to review modern approaches for their practical implementation.
The objectives of the article are to explore the theoretical aspects of privacy. Analyze security policies and models, threat models, and security criteria. To investigate each of the component criteria for the security of complex information systems. To carry out the threat modeling process using the STRIDE methodology. To evaluate the specified methodology and perform calculations to assess the confidentiality of the system.
Methods
Mathematical and computer modeling. Statistical analysis. Comparative analysis. Literature analysis. Generalization and systematization of the material.

Keywords: confidentiality, violation of confidentiality, methods of ensuring confidentiality, generative multimodal system, threat model, security model, threat criteria, unauthorized access, STRIDE, security assurance

In recent years, the safe operation of energy facilities has increasingly been ensured by probabilistic non-destructive testing systems. This article examines a method for predicting and estimating the number of missed defects by solving an inverse problem. A detailed analysis of indirect manifestations and prediction of an indirect parameter is conducted using the Keras deep learning library, which determines the quantitative characteristics of the facility under study. The results of the study demonstrate encouraging prediction accuracy with easily correctable signs of model overfitting.

Keywords: non-destructive testing, defects, defect detection probability distribution curves, synthetic data for deep learning, regression forecasting, Keras, structural and semantic features, non-linear dependencies

In this paper, we propose a method for evaluating the key indicators of a multichannel queuing system with an unlimited queue and multiphase Erlang-type service. It is shown that the transition to the multichannel case leads to a sharp increase in the dimension of the state space and a complication of the system of Kolmogorov equations, which often makes direct analytical calculation unavailable. A meta-model based on machine learning methods, trained on discrete event simulation data, is proposed for an approximate forecast of the average waiting time, average queue length and the proportion of applications served. A comparison of basic regression and neural network models is performed and the stability of the approximation with a change in the load factor is considered.

Keywords: queuing system, queue, simulation modeling, meta-model, machine learning, neural network, multi-channel service, Erlang distribution, impatience, Kolmogorov equation, regression, gradient boosting, random forest, perceptron

This paper concerns the problems of identification of systems with mixed-type nonlinearities. An improved method of frequency identification using a system of correlators, which allows recording a bilinear frequency response is shown. An approach is proposed that provides a more accurate measurement frequency response of Volterra kernels, which consists of output correction. The efficiency of the proposed approach is demonstrated using the example of a nonlinear system including a deadband block. Based on known analytical values, the errors of the methods are calculated.

Keywords: Volterra series, system identification, nonlinear systems, piecewise nonlinearities, frequency responses, Riccati equation

The study is devoted to the analysis of modern approaches to the organization of body scanning processes using photogrammetric technologies. Various methods of digital reconstruction of the human body are being considered, including manual and robotic scanning systems. A comparative analysis of measurement accuracy, texture quality and time characteristics of various approaches was carried out. Particular attention is paid to the issues of rigidity of the structures of scanning systems and their impact on the quality of the resulting three-dimensional models. The results show the superiority of specialized hand-held scanners over photogrammetric methods in accuracy, but the greater versatility of the latter in various application conditions.

Keywords: photogrammetry, 3D scanning, digital reconstruction, measurement accuracy, structural rigidity, robotic systems

The article presents a hybrid neural network for estimating the mass of a car and the longitudinal/transverse slopes of a road, combining a square-root sigma-point Kalman filter and a neural network model based on a transformer encoder using cross-attention to the evaluation residuals. The proposed approach combines the physical interpretability of the filter with the high approximation capability of the neural network. To ensure implementation on embedded electronic control units, the model was simplified by converting knowledge into a compact network of long-term short-term memory. The results of experiments in various scenarios showed a reduction in the average error by more than 25% with a computational delay of less than 0.3 ms.

Keywords: vehicle condition assessment, road slope assessment, vehicle mass assessment, transformer neural network, cross-focus, adaptive filtering, knowledge distillation, square-root sigma-dot Kalman filter, intelligent vehicles, sensor fusion

The article analyzes the main types of distributed denial-of-service attacks and explores classical and innovative methods of protecting web servers from threats, including packet filtering, intrusion detection and prevention systems, and load balancing architectures. Based on the research results, significant limitations of traditional approaches have been identified, such as low adaptability to new threats, high false positive rates, and inability to effectively counter modern multi-factor attacks. The paper highlights the potential of using artificial intelligence and neural networks to analyze network traffic and detect complex patterns of anomalies.

Keywords: web server protection, distributed attack, denial of service, traffic filtering, packet filtering, intrusion detection system

The article discusses the issues of assessing the robustness of memristor elements in order to increase the reliability of artificial intelligence systems based on nanoelectronic structures. The system of nonlinear equations describing the frequency response of a memristor, the input signals of which can strongly depend on various parameters, cannot be solved using methods accessible to standard mathematics.
To achieve the result, it is proposed that the system of equations be solved using interval arithmetic methods. The value of intermediate solutions lies in the fact that they provide access to the most reliable solutions to basic problems, taking into account possible changes in the initial and calculated values.
The main task of interval computing is to replace arithmetic operations and real functions on real numbers with interval operations and functions that transform intervals containing these numbers. In interval calculation, the main object of research is the interval, which is a closed numerical interval. The value of interval calculations lies in the fact that they contain accurate solutions to the initial problems. The interval calculation methods developed to date are based on the use of arithmetic operations with real and complex numbers.
Using interval calculation can help reduce errors in calculations and data storage in electronic devices. For example, when using memristors to store information, interval calculation can help account for factors affecting data read and write errors. Interval arithmetic allows you to take into account possible errors and uncertainties that may arise during measurements and calculations. This helps to reduce the likelihood of errors and increase the accuracy of forecasting the operation of memristors.
The proposed algorithm for estimating the robustness of a memristor in the mode, which takes into account significant increases in the nonlinearities of electrical parameters from the point of view of reliability, makes it possible to calculate the characteristics of the developed circuits and reduce the time spent on circuit engineering when searching for the best option.

Keywords: memristor, multipole, topological graph, finite increments, structural-parametric model, algorithm

An algorithm for modeling smooth hysteresis nonlinearities is proposed, taking into account the slope coefficient k and the saturation level c. The developed model provides accuracy and ease of adjustment while maintaining intuitive physical parameters of the hysteresis loop, which makes it effective for practical application in the tasks of analysis and synthesis of nonlinear control systems.

Keywords: unambiguous nonlinearities, hysteresis, automatic control systems, backlash with saturation, ambiguous nonlinearities, algorithm, modeling of automatic control systems, relay, static characteristic, approximation

The article discusses integrated development environments based on artificial intelligence as an innovative programming tool that provides automation of routine software development tasks. The Cursor development environment, developed by Anywhere, is the main object of research. The architectural features of the system are analyzed, including an agent-based approach to interacting with code, context management mechanisms through generation supplemented with extracted data, and code base indexing using vector representations and Merkle trees to optimize updates. The key limitations of modern integrated development environments based on artificial intelligence have been identified: problems with the size of the context window, indexing performance of large repositories, accuracy of context extraction, as well as privacy and security issues. Special attention is paid to the human factor – the lack of competence of developers in the field of effective context management and the creation of high-quality products. The article substantiates the need to create a preliminary context management agent capable of technically optimizing processes and directing users to effective practices of working with integrated development environments based on artificial intelligence.

Keywords: integrated development environment, artificial intelligence, Cursor development environment, large language models, context management, generation with addition of extracted data, code base indexing, Merkle trees, agent-based approach, software developmen

A reproducible method is presented for the autonomous determination of the coordinates of the base stations of fourth-generation mobile radio networks and the parameters of their sectors, based solely on field observations of the modem without using time delay and methods for estimating the angles of arrival of the signal. The approach combines robust allocation of an informative "core" of measurements, weighted minimization of distances and aggregation at the site level, providing stable estimates in urban environments. Experimental verification in a real scene demonstrates a significant reduction in localization error compared to the basic centroid and median methods, which confirms the practical applicability of the proposed solution.

Keywords: LTE, positioning, localization, base station, site coordinate, signal strength, angular distribution, sectorization, optimization, observation, geometric median, field recording, minimization method, radio signal

The work compares the use of recurrent networks and models based on transformer architecture to solve the problem of predicting the completion time of a business process. Models, by definition, model the sequence of actions and are able to take into account a variety of attributes in determining target characteristics. For comparison, a recurrent model of long-term short-term memory and a transformer encoder of its own architecture were used, the operation of which was tested on openly presented real data from the logs of the support service. The models were trained and tested using Python using the pandas, numpy, and torch libraries with the same data preparation, prefix generation, and time division for both models. The comparison as a result of experiments on the average absolute error showed the advantage of the transformer encoder; approximately the same accuracy of the models with a slightly higher accuracy of the transformer model was recorded for the standard deviation.

Keywords: predictive monitoring, event log, machine learning, transformer encoder, neural networks, data preparation, regression model, normalization, padding, recurrent network, model architecture

The article presents the results of calculations of the stress-strain state of the Sophie Germain-Lagrange plate obtained using approximate methods for solving differential equations: the Bubnov-Galerkin method, the finite difference method and the differential quadrature method. It is shown that the differential quadrature method using the Chebyshev grid is an effective method for solving bending problems of thin rectangular plates and allows obtaining high-precision results using a limited number of nodes.

Keywords: differential equation, approximate solution, differential quadrature method, Chebyshev grid, Sophie Germain-Lagrange plate, plate bending, stress-strain state

The article addresses the solution of systems of linear algebraic equations arising in static and steady-state vibration problems solved by the finite element method. A block-sparse matrix format based on the CSR (Compressed Sparse Row) format is presented, along with its GPU implementation using CUDA. A stabilized biconjugate gradient method is implemented and applied to model problems of varying dimensions; a comparison with a reference implementation in MATLAB is also conducted.

Keywords: sparse matrices, finite element method, block matrices, GPU, parallel computing, systems of linear algebraic equations, biconjugate gradient method

In the context of the increasing complexity of multi-storey construction and the need to ensure its safety, reliability and economic efficiency, the issue of improving construction control is of critical importance. This article explores the current problems of traditional methods of construction control (subjectivity, reactive nature, risks of non-compliance with comfort and engineering) and suggests comprehensive ways to overcome them through the integration of modern digital technologies.
The main focus is on a proactive customer control model based on the integration of information modeling technology, three-dimensional scanning (3D scanning) and artificial Intelligence technologies, which ensures continuous verification of compliance with the "as built" design solution.

Keywords: construction control, innovation, verification, information model, technology, digitalization, three-dimensional scanning, energy efficiency

The article presents the architecture and implementation of an intelligent software package (ISP) for predicting the thermal resistance of semiconductor devices, in particular MOSFET transistors, at the design stage. The developed system combines physical and mathematical modeling of multilayer heat-conducting structures with machine learning methods, which allows for an accurate prediction of thermal parameters based on engineering characteristics and case design. The ISP implements a mechanism for automatically supplementing incomplete data using a knowledge base of typical parameters of domestic and foreign devices. Models were trained on a synthetically expanded sample formed taking into account the thermal conductivity of structural materials and layer geometry. Among the algorithms used are ensembles of random forests and gradient boosting, as well as neural network models. An analysis of the importance of features was carried out, key parameters that determine them were identified, and the possibility of using the ISP for early assessment of thermal conditions in CAD and CAE environments was demonstrated.

Keywords: thermal resistance, MOSFET, machine learning, intelligent software system, multilayer structure, predictive model, CAD, thermal conductivity

The paper examines the nature of exponential behavior and identifies the conditions under which the probabilistic distribution of the project completion period deviates from the exponential one. For this purpose, a model has been developed in which the evolution of the project is described as a Markov process with a transition matrix containing a constant in all elements of the first row. This structure corresponds to a situation in which the project can be restarted at any time. Project completion times can follow various statistical distributions, including normal, exponential, and more complex forms. Examples of such projects can be research, exploration, venture and other similar projects. An analysis of the dynamics shows that the model reliably reproduces the exponential distribution in cases where the probability of a restart remains moderate. This indicates the limit of applicability of the exponential description.: it is adequate for low and medium restart probabilities, but loses accuracy with a high level of uncertainty.

Keywords: Markov processes, project management, exponential distribution, project completion time, risk assessment, probabilistic forecasting, uncertainty in projects, risks of assumptions, dynamics of project evolution

Construction work often involves risks when carrying out complex sets of tasks described in the form of network schedules, in particular, risks of violating tender deadlines and project costs. One of the main reasons for increased project risks is a lack of resources. The main objective of this study is to develop a methodology for modeling network schedules under resource constraints, taking into account the stochastic influence of risks on project completion deadlines. The paper analyzes tools for modeling project schedules; describes a mathematical model for estimating project cost based on a network schedule under resource constraints; proposes a method for modeling a network schedule in the AnyLogic environment; develops an algorithm for modeling parallel branches of a project schedule under resource constraints; and describes a method for modeling a network schedule for project work. Testing was conducted based on a network schedule for a project to construct a contact line support. It has been shown that the method allows for obtaining probabilistic estimates of project deadlines and costs under conditions of risk and limited resources. The methodology can be applied to various projects described by network schedules and will allow solving a number of practical tasks: optimizing resources allocated for project implementation, taking into account the time and cost of the project, analyzing risks affecting project implementation, and developing optimal solutions for project risk management.

Keywords: network schedule, work plan, simulation modeling, risk analysis, project duration, project cost

The paper discusses issues related to the use of graph-theoretic models in text analysis. One of the tasks is to aggregate such models to identify more "simple" graphs, the vertices of which correspond to subsets of the vertices of the original model, and the edges reflect the "strong connections" between the vertices. Using the example of a Russian folk tale plot, it is shown how to build an aggregated model with a given threshold of significance and present it for further analysis. To conduct the experiments, a set of graph-theoretical models for fairy-tale plots from the collection of A.M. Afanasyev was built using the Folklore information system, where the graph aggregation module was improved.

Keywords: text analysis, graph-theoretical model, aggregation, significance threshold, storage format, folklore text, fairy tale plot, information system "Folklore"

The article deals with the problem of quantifying the security of objects of the penal correction system based on mathematical modeling. The authors propose a modified genetic algorithm in which the traditional fitness function is replaced by a mechanism of "virtual movement" of individuals in a discrete space, which allows taking into account both the individual characteristics of safety measures and their cumulative impact. A step-by-step example of applying the algorithm to choosing the optimal set of protective measures based on expert assessments based on several criteria is given. The results demonstrate the effectiveness of the proposed approach for solving multi-criteria tasks of assessing and improving security in conditions of uncertainty and the absence of an explicit analytical relationship between system parameters.

Keywords: security, vulnerability, event, penal enforcement system, genetic algorithm, mathematical modeling, optimization, expert assessment, criterion, security, security regime, discrete space, coalition, crossing, fitness function

The paper presents an analysis of third-generation spiking neural networks application to solving regression tasks. It considers the main models of spiking neurons (LIF, Izhikevich, Hodgkin-Huxley) from the perspective of their computational complexity and suitability for regression problems. Methods for encoding real-valued data into spike sequences are analyzed: rate coding, temporal coding, and population coding. Special attention is given to methods of decoding output spikes into continuous values, including rate decoding, first spike timing decoding, membrane potential decoding, and population voting. An assessment of the energy efficiency of various approaches is conducted, demonstrating a 100-200 fold reduction in energy consumption compared to traditional neural networks while maintaining acceptable accuracy. The research results confirm the promising application of spiking networks in embedded systems and Internet of Things devices.

Keywords: spiking neural networks, spike neuron model, spike coding, regression, energy efficiency

The article presents the results of a study of the security of the command transmission channel for unmanned aircraft (UAV) using the example of an FPV drone. The research was carried out in an anechoic shielded chamber of a specialized landfill with certified measuring equipment. The results of measurements of the spectral panorama and the possibility of passive interception of signals in the radio are presented. The relevance of ensuring safe operation of the UAV is shown, as well as the vulnerability of the ELRS protocol to control interception. Recommendations on the use of cryptographic algorithms to neutralize security threats are given.

Keywords: unmanned aircraft, UAV, FPV drone, ExpressLRS, FHSS, SDR receiver, safe operation, information protection, vulnerabilities, unauthorized access, control interception, identification phrase

The paper investigates the specifics of the digital implementation of a rotor synchronization control algorithm for the SV-2M two-rotor vibration unit. The influence of the sampling period on the stability of the synchronous mode is evaluated using computer simulations in MATLAB/Simulink, with time quantization and zero-order hold delay explicitly accounted for. A comparative analysis of the digital and analog versions of the algorithm has been performed. The boundary values of the discretization periods for different values of the given total energy of the system have been determined. The obtained results confirm the applicability of the proposed approach for digital control systems of vibratory equipment.

Keywords: digital control, rotor synchronization, two-rotor vibration unit, speed-gradient algorithm, extrapolator

This article examines transportation network modeling using the Ford–Fulkerson algorithm. It describes the process of finding a minimum cut using a graphical editor and library developed in the C# programming language. Key concepts of graph and network theory are presented to clarify the problem statement. An example of solving a transportation problem using the developed software is shown, and the program's results are compared with a control example.

Keywords: transportation network, maximum flow problem, Ford–Fulkerson algorithm, minimum cut in the network, software library, graphical editor, C# programming language