Vol 26, No 1 (2025)

The article explores the potential of applying machine learning (ML) for adaptive trajectory control of unmanned aerial vehicles (UAVs) under uncertainty. The concepts of ML algorithms and the classification of UAVs by purpose, size, and weight are examined. To analyze control methods, theoretical approaches such as ensemble learning, neural networks, and probabilistic models are applied, enabling real-time adaptation of flight trajectories. Additionally, mathematical models are presented and illustrated with formulas describing the dynamics of interaction between the control system, external disturbances, and control inputs. Parameters such as system adaptability, trajectory correction accuracy, and stability under challenging conditions are studied to assess the accuracy and efficiency of the proposed algorithms. The study also investigates the impact of computational power limitations on the real-time performance of algorithms. The integration of data from various sensors is considered crucial for improving the accuracy and reliability of the control system. Special attention is given to the practical application of ML for environmental change prediction and flight trajectory optimization. Examples of real-world ML algorithm implementations include successful developments by Russian and foreign companies, demonstrating high levels of autonomy and adaptive control. The results show that ML significantly enhances UAV autonomy and safety, ensuring reliable trajectory corrections even under uncertain conditions. Further research could focus on developing collective control for UAV groups and improving real-time ML integration. This would expand UAV functionality, improve efficiency, and reduce resource consumption.

In the context of integrating wind farms into power systems, it is imperative to ensure a reliable power supply to consumers. Changes in the operating modes of wind power plants (WPPs) should be compensated for by the ability of the control range of conventional power plants to load or unload active power. Therefore, when increasing the installed capacity of WPPs in power systems, improving the manoeuvrability characteristics of thermal power plants, including the expansion of the regulation range, is the main condition for reliable operation of the power system. The results of research on calculation of dynamic non-uniformity coefficient for different power systems with wind power plants are presented. Comparison of the results made it possible to establish that changes in WPP power with an amplitude of up to 40% of the installed or baseline WPP power and a fluctuation period of 15 minutes to 3 hours constitute the main time duration (about 90%). Using the Australian power system as an example, it is shown that the distribution of WPPs across the power system has a positive effect on load schedule levelling.

A new mathematical model is presented aimed at improving the accuracy of forecasting and the effectiveness of managing organizational projects in conditions of uncertainty. The relevance of the study stems from the growing complexity of projects and the need to develop tools to minimize risks and optimize resources. The aim of the study is to create a mathematical apparatus that allows assessing the influence of various factors on the dynamics of the project and developing management mechanisms that are resistant to external influences. To achieve this goal, an analysis of existing approaches to project management was conducted, a mathematical model was developed based on systems theory and control theory. A systematic approach was used to build the model, which helped to account for the interrelationships between the various elements of the project. The model takes into account such factors as uncertainty, resource constraints and process dynamics. The results of the study allowed us to create a tool for forecasting and optimizing project management. The model helps to identify potential risks and develop effective strategies to minimize them. The scientific novelty of the work lies in the development of a new mathematical model that takes into account the specifics of organizational projects in order to accurately predict their results. The practical significance of the study lies in the possibility of using the developed model to improve the efficiency of project management in various organizations.

This study explores the application of modern data processing methods - wavelet transformation, stochastic methods, and Support Vector Machine (SVM) - on real electroencephalogram (EEG) signals from open databases. Analyzing EEG signals is crucial for medical diagnostics and neuroscience, requiring sophisticated techniques due to high dimensionality and noise. Wavelet transformation allows decomposition of signals into frequency components with varying temporal resolutions, facilitating time-frequency analysis. Stochastic methods utilize probabilistic models for modeling random processes and analyzing data statistics. Meanwhile, SVM is a machine learning algorithm that identifies the optimal hyperplane to separate classes, enhancing generalization, particularly with complex nonlinear data. When comparing these methods, the specific data type and task should be considered: wavelet transformation is ideal for signal processing, stochastic methods are used for random processes, and SVM excels in classification tasks. Thus, selecting the most suitable approach should be based on a comparative analysis of method effectiveness in a particular context. This study will discuss these concepts and present examples of applying these techniques to EEG data, contributing to the analysis and classification of brain activity and the identification of pathologies.

One of the biggest challenges that engineers encounter in a variety of industry sectors is corrosion. The current research focuses on the corrosion behavior of various types of aluminum alloys widely used in the industry. In this regard, aluminum alloys Al2024, Al6061, and Al7075 were tested. Also, the effect of environmental temperature on the corrosion rate of each group of materials was investigated. Three statistical parameters, including total corroded area, corrosion rate (total corroded area to total sample area), and the maximum size of corroded point, were measured as corrosion indicators in the samples. In addition, the surface hardness of the samples was measured and presented by the Brinell method. Finally, the weakest aluminum alloy against corrosion under different temperature conditions was introduced. The corrosion test conducted in the presence of cold air produced the maximum hardness in any of the aluminum alloys (2024, 6061, and 7075) that were examined. Aluminum 7075 has the lowest corrosion resistance, while aluminum 6061 has the strongest corrosion resistance when various testing conditions are taken into account.