This article addresses calculation of the thermophysical processes occurring in a thermoelectric device (TED) for the correction of cerebral hyperthermia by cooling the head. The computational model consists of solutions to three problems: determination of the temperature field of the head area undergoing cooling; determination of the parameters of the thermoelectric modules (TEM) included in the TED; determination of the characteristics of the heat removal system from the hot junctions of the thermal elements in the thermal modules.

This work explores the mathematical modeling of thermophysical processes within a thermoelectric device (TED) designed for brain cooling. The pressing need to mitigate cerebral hyperthermia, a condition triggered by factors such as strenuous physical activity, elevated ambient temperatures, and various pathological conditions, underscores the significance of this research. The focus of this study is a specialized design—a cooling helmet encompassing the scalp, frontal, and temporal regions. This helmet interfaces with the head through a block of thermoelectric modules (TEM) to facilitate cooling. The mathematical model developed for this research addresses three critical components (i) temperature field calculation, (ii) TEM parameter optimization, and (iii) heat dissipation system. The outcomes of this mathematical modeling endeavor yield temperature profiles at key control points within the TED, depicted in spatial and temporal coordinates, contingent upon the supplied TEM power. Furthermore, the study uncovers essential energy parameters for the thermomodules, encompassing their particular types sourced from the standard TEM selections offered by “Cryoterm” LLC, the manufacturer. Notably, the theoretical investigation into TEDs for brain cooling unveils a pragmatic remedy. The integration of two typical DRIFT-0.8 TEMs from the specified manufacturer within the TED design demonstrates the capacity to reduce brain temperature to 291 K. The characteristics of these TEMs display variation within predefined thresholds, with power outputs spanning from 20 to 80 W, an average temperature difference of 50 K across junctions, supply currents ranging from 3 to 11 A, and power consumption levels ranging from 100 to 500 W. These attributes are complemented by coefficient of performance values ranging from 0.1 to 0.5. This research not only sheds light on the feasibility of employing thermoelectric devices for brain cooling but also lays the foundation for addressing cerebral hyperthermia and advancing neuroprotection applications across various clinical and non-clinical contexts.

The theoretical possibilities of increasing the resolution and sensitivity of a time-of-flight mass spectrometer with orthogonal ion injection are considered. The effects are achieved by using inhomogeneous electrostatic fields of special configurations both in the accelerating and focusing parts of the device – a cylindrical immersion objective and a transaxial mirror, respectively. It is shown that the use of an inhomogeneous cylindrical field of a special configuration as an ion accelerator opens up the possibility of a multiple reduction in the energy spread of ions in injected ion packets, associated with the so-called "turnaround time" and, therefore, a significant (two or more times) increase in the limiting resolution of the mass spectrometer. The use of a transaxial electrostatic mirror as a time-of-flight mass analyzer makes it possible to significantly increase the sensitivity of the mass-spectrometer due to the implementation of triple space-time-of-flight focusing of ion packets. Key features include reduced ion energy spread, increased maximum resolution, and improved sensitivity due to triple focusing in space and time of flight. This research lays the foundation for expanding the capabilities of time-of-flight mass spectrometry, providing a more efficient and powerful tool for a wide range of scientific and industrial applications. The effects are achieved by using inhomogeneous electrostatic fields of a special configuration in both the accelerating and focusing parts of the device – a cylindrical immersion lens and a transaxial mirror, respectively. Numerical calculations of the system – a four-electrode cylindrical immersion lens in combination with a three-electrode transaxial mirror – are presented, which confirm the conclusions of the theory

Abstract This paper presents a comprehensive study aimed at improving the efficiency of unmanned aerial vehicles (UAVs) through the enhancement of their aerodynamic and mechanical structures. The research is based on coupled computational fluid dynamics (CFD) and finite element analysis (FEA). The airflow around the UAV was modeled using the Navier–Stokes equations, while the structural behavior was described by the equations of linear elasticity. A UAV configuration with a wingspan of 1.8 m and a mass-optimized structure was investigated for flight speeds in the range of 10–35 m/s and angles of attack from −5° to +15°. The results of the aerodynamic optimization, including airfoil thickness variation and smoothing of the wing–fuselage junction, showed a reduction in the drag coefficient by 9–12% and an increase in the lift-to-drag ratio by up to 11% in the cruise regime. The structural optimization based on replacing aluminum with a carbon-fiber composite material led to a reduction in the structural mass by 13–16%, a reduction in the structural strength criterion value by 18–22%, as confirmed by the Tsai–Wu failure analysis, and a reduction in wing-tip deflection by 20–25% under 3 g and 5 g load cases, while satisfying strength and stiffness requirements. The obtained results demonstrate that the proposed integrated aerodynamic and structural optimization approach significantly improves the overall performance, efficiency, and operational reliability of UAV systems. © 2026 by the authors.

This paper presents a large-scale empirical study aimed at identifying the optimal local deep learning model and data volume for deploying intrusion detection systems (IDS) on resource-constrained IoT devices using federated learning (FL). While previous studies on FL-based IDS for IoT have primarily focused on maximizing accuracy, they often overlook the computational limitations of IoT hardware and the feasibility of local model deployment. In this work, three deep learning architectures—a deep neural network (DNN), a convolutional neural network (CNN), and a hybrid CNN+BiLSTM—are trained using the CICIoT2023 dataset within a federated learning environment simulating up to 150 IoT devices. The study evaluates how detection accuracy, convergence speed, and inference costs (latency and model size) vary across different local data scales and model complexities. Results demonstrate that CNN achieves the best trade-off between detection performance and computational efficiency, reaching ~98% accuracy with low latency and a compact model footprint. The more complex CNN+BiLSTM architecture yields slightly higher accuracy (~99%) at a significantly greater computational cost. Deployment tests on Raspberry Pi 5 devices confirm that all three models can be effectively implemented on real-world IoT edge hardware. These findings offer practical guidance for researchers and practitioners in selecting scalable and lightweight IDS models suitable for real-world federated IoT deployments, supporting secure and efficient anomaly detection in urban IoT networks. © 2025 by the authors.

The object of this study is to robotize the technological operation of removing the oxide film from the surface of a magnesium melt poured into continuously moving molds of a casting conveyor for the production of commercial magnesium. To robotize this technological operation, it is proposed to use a two-armed manipulation robot with a spherical coordinate system, which has six degrees of mobility. Software trajectories have been developed according to the degrees of mobility of the manipulation robot in terms of position, speed, and acceleration to perform the technological operation of removing the oxide film from the surface of the magnesium melt poured into the moving molds of the foundry conveyor. Programmed trajectories are described by quadratic polynomials that satisfy restrictions on the values of the generalized coordinate, velocity, and acceleration. These limitations are determined by the design features and energy capabilities of the degrees of mobility drives of the manipulation robot. Programmed trajectories along the first and second degrees of freedom compensate for the continuous movement of the molds of the foundry conveyor. Programmed trajectories along the third and fourth degrees of mobility enable the collection of the oxide film from the surface of the magnesium melt. Programmed trajectories along the fifth and sixth degrees of freedom enable the discharge of the collected oxide film into a special container. The reliability of the developed programmed trajectories is confirmed by the simulation results using MATLAB version R2015b. Based on the results, a cyclogram for controlling a manipulation robot has been constructed to perform the technological operation of removing the oxide film in the production of commercial magnesium. The results could be used in the robotization of technological processes for removing the oxide film in the production of commercial magnesium or similar foundries. © 2024, Authors. This is an open access article under the Creative Commons CC BY license

This paper was performed using geometry and interactive motion (GIM) software, a learning and research software designed to facilitate the kinematic analysis and synthesis of planar mechanisms. This study presents an analytical and geometric approach to the synthesis of planar four-bar linkages for precise path generation using sixth order connecting coupler curves. Emphasis is placed on defining linkage configurations that enable a coupler point to pass through a set of three, four, or five user-defined precision points. The methodology integrates classical kinematic principles, Roberts' theorem, and focal circle theory to identify double points and optimize the trajectory of the coupler. Using complex number representations and algebraic formulations, the equations governing coupler motion are derived and applied to various linkage types, including crank-rocker, double-rocker, and limit mechanisms. Parallelogram constructions and focal circles are used to detect critical configurations and analyze the continuity and self-intersections of coupler paths. Computational simulations validate the synthesis approach across different configurations, highlighting trade-offs between accuracy and mechanical feasibility. The results demonstrate the applicability of the method in advanced motion control tasks such as robotic arms, dwell mechanisms, and function generation. This work contributes to the field of kinematic synthesis by offering a rigorous framework for linkage design with enhanced precision, reliability, and versatility. © The Author(s) 2026. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits the use, sharing, adaptation, distribution and reproduction in any medium or format, as long as appropriate credit to the original author(s) and the source is given by providing a link to the Creative Commons license and changes need to be indicated if there are any. The images or other third-party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

The object of this study is the technological operation of removing the oxide film from the surface of the metal melt, foundry production of commercial lead, zinc. To carry out the robotization of this technological operation, it is proposed to use a manipulation robot with a spherical coordinate system. A kinematic structure of a manipulation robot with six degrees of mobility and two arms is proposed. On the first arm of the manipulation robot, a movable blade is fixed, and on the second arm, a rotary blade is fixed. With the translational movement of the first hand, the movable blade rakes the oxide film onto the rotary blade. Further, the oxide film collected on the rotary blade is thrown into a special container with a rotational movement. Restrictions are introduced on the values of generalized coordinates, velocities, and accelerations for each degree of mobility of the manipulation robot. Taking into account these limitations, for the implementation of this process, software trajectories have been developed for the degrees of mobility of the manipulation robot, which are approximated by quadratic polynomials. Each program movement is divided into three sections, in the first section acceleration with a given acceleration is carried out, in the second section movement with a given speed, in the third section braking with a given acceleration. To assess the reliability of the developed software trajectories, simulations were carried out in the MatLab software environment, version R2015b. The resulting graphs of program trajectories coincide with the calculated values of the generalized coordinates, time intervals, speeds, and accelerations of change in the generalized coordinates in terms of the degrees of mobility of the manipulation robot. The period of time required to remove the oxide film is 15.88 s. On the basis of the results obtained, a cyclogram for controlling a manipulation robot was built to perform the technological operation of removing the oxide film in the production of commercial lead, zinc

Abstract. The task of controlling multi-channel objects, that is, objects with many inputs and many outputs, MIMO, can be the most effectively solved by a multi-channel controller that has the same number of inputs and outputs as the controlled object has. We consider only multichannel controllers that have the same number of inputs and outputs. Traditionally, in the matrix of such a controller, each element is a scalar PID controller, that is, a controller containing proportional, integrating and derivative channels. In this case, an n × n object requires n 2 scalar PID controllers, each with three channels. The calculation of such a controller requires the calculation of 3n 2 parameters.

Целью данного исследования является применение методов искусственного интеллекта (ИИ), в частности искусственных иммунных систем (ИИС), для разработки оптимальной стратегии управления многопараметрической системой управления. Исследуются два конкретных подхода к управлению в промышленности: интегрально-пропорционально-дифференциальное (ИПД) и пропорционально-интегрально-дифференциальное (ПИД) управление. Мотивацией для использования этих вариантов ПИД-регуляторов является их функциональная реализация в современных промышленных контроллерах, где они обеспечивают точное управление технологическим процессом. Результаты исследования представляют собой новое решение задачи синтеза управления для промышленной системы.