Ensuring the security of data transmission in agrotechnical activities is crucial, especially when using advanced monitoring systems based on UAVs and AI methods. Traditional encryption methods face significant threats due to the emergence of quantum computing. This study explores the application of Quantum Key Distribution (QKD) to secure data transmission in UAV-based geographic information systems (GISs) used for monitoring both forest fires and agricultural fields. By leveraging the BB84 protocol with polarization of weak coherent pulses, quantum keys are distributed between UAVs and ground stations, ensuring data integrity and security. The hardware requirements for integrating QKD in UAVs and ground stations include compact lasers, polarization modulators, microlenses, polarization filters, and single-photon detectors. Simulation results indicate that the key generation speeds are sufficient for real-time secure data transmission, even under the constraints of UAVs such as limited power and size. Furthermore, this study examines the influence of atmospheric conditions, geometric losses, and receiver characteristics on the communication range and stability. The proposed QKD method enhances the efficiency of data security in GISs for agricultural production monitoring. Future research will focus on practical implementation, optimization of the QKD system for UAVs, and integrating QKD with existing communication systems and data transfer protocols. This integration aims to provide robust support system for agrotechnical activities, leveraging advanced AI methods and monitoring systems to increase the efficiency and security of agricultural production.

Frequency-follower robots are devices designed to detect, identify, locate, or follow radiofrequency signals. These robots are equipped with antennas and receivers calibrated to receive specific frequency ranges. Many of these robots are designed for specific purposes, such as identifying interference frequencies or analyzing the coverage of a transmitter. In this context, to provide a versatile solution for tracing frequency signals in the field or remote areas, this work presents the development of a small, easily implemented robotic vehicle that facilitates the identification of specific, pre-calibrated radio frequencies. The robot is designed with a four-wheeled configuration similar to a standard line-follower robot, allowing for stable and precise movement. It is equipped with a controlled receiver and a microcontroller that processes incoming signals and directs the robot’s movements accordingly. This setup enables the robot to autonomously navigate and trace frequency signals, even in challenging environments. To ensure the robot’s effectiveness, a thorough characterization process was conducted. This involved calibrating the detection frequencies and fine-tuning the robot’s trajectory-tracing capabilities. The robot was tested in various field conditions, including areas with difficult access, to evaluate its performance. The characterization process confirmed the robot’s ability to detect frequencies in the ranges of 10 to 100 MHz, 10-50 MHz, and 200-500 MHz. Additionally, the robot demonstrated its capability to track frequency coverage in peripheral and high-mountain areas for its versatility and adaptability.

Power quality (PQ) measures system reliability, equipment security, and power availability in electrical power systems. Common PQ problems include voltage sags, swells, overvoltages, undervoltages, harmonics, transients, and grounding issues, with harmonics and sags having the most significant impact. PQ events are variations in voltage magnitudes and waveform distortions affecting low-frequency to high-frequency spectral content deviations and other phenomena. Power quality is a growing concern due to the restructuring of the electric utility industry and the proliferation of small- and medium-scale distribution generations. This study investigates the predictive monitoring and analysis of power quality (PQ) events in a three-phase microgrid using digital signal processing (DSP) and machine learning (ML) approaches. The results show that harmonics and voltage sags are prevalent issues in AC microgrids, affecting system stability and equipment performance. This study also compares the microgrid's performance with the utility grid, showing that converter-interfaced systems are more susceptible to harmonics combined with RMS voltage variations. The microgrid showed lower Total Harmonic Distortion (THD) but increased sensitivity to voltage sags, highlighting the need for careful consideration when operating high-power devices. This research emphasizes the importance of PQ monitoring in power systems and the significance of both long-term and short-term monitoring for effective power system operation and equipment safety.

The rapid development of intelligent transportation systems (ITSs) has made vehicular networks (VNs) indispensable, particularly through vehicle-to-everything (V2X) communication. This study proposes an advanced framework for the construction and migration of digital twins (DTs) in vehicular networks to improve decision-making and predictive maintenance. The construction phase utilizes a large model-driven framework enhanced by an advanced deep reinforcement learning (DRL) algorithm, specifically an Enhanced Deep Q-Network (EDQN). This framework processes complex and dynamic vehicular data, supporting EDQN in optimizing decision-making processes. EDQN adapts dynamically to vehicular environments, ensuring high decision accuracy and efficiency. In the migration phase, due to limited base station coverage, transfer learning techniques are employed to enable the seamless migration of DTs across different base stations. This method minimizes computational overhead compared to traditional approaches by adapting pre-trained models to new environments with minimal retraining. Experimental simulations demonstrate that the integration of the large model architecture with EDQN significantly enhances decision-making processes. The transfer learning strategy effectively extends the operational coverage, maintaining high performance and service continuity during DT migration. This research underscores the potential of leveraging advanced AI techniques to improve the management and operational efficiency of vehicular networks, providing a robust foundation for future advancements in ITS.

A power distribution company, along with any other company that consumes a significant amount of energy, has a substantial demand for dependable power to earn income and produce goods. According to the research findings, transformers are precious assets for businesses; hence, the maintenance and replacement of transformers are considered a luxurious activity for every business institution. Considering the factors above, this work develops an IoT-Driven Distribution Transformer health monitoring system for smart power grid. Remotely monitoring the health of distribution transformers at predetermined intervals is the objective of this system. Changes in current values on phases, phase failure, overvoltages, overcurrent, earth fault, undervoltages, oil temperature, body temperature, and load ability are just a few of the factors that are used to calculate the health index. Sensors detect these factors. It has been decided that Arduino will serve as the processor for transferring the data that has been sensed, and that the Blynk App will serve as the Internet of Things platform for presenting the data that has been received. The installation process occurs in the physical vicinity of the distribution transformer. After being processed, the values output by the sensors are stored in the system's memory. The system has predefined instructions to detect abnormal situations, which are automatically updated via serial communication on the internet. It is possible to put this low-cost technology in transformers anywhere so they can be monitored remotely. This not only helps assess the health state of the transformers but also assists in projecting their anticipated lifespan. The Internet of Things (IoT) can optimize transformer usage and detect potential issues before catastrophic collapse. An online-measuring system collects and analyzes data on oil temperature, body temperature, voltage, and current, enabling Transformer Health Measuring to identify unforeseen circumstances, resulting in increased reliability and cost savings.

Railway signaling is a critical component of ensuring the safe and efficient operation of train systems. The preparation of signaling control (interlocking) tables, which outline a specific signaling logic for controlling signals, switches, and track section vacancy, can be a complex and time-consuming task. These tables involve intricate logic, considering train positions, speeds, and potential conflicts within each possible train route. The generation of control tables for small railway stations can be performed manually relatively quickly, but the complexity of the work grows exponentially with the topology of the corresponding railway station, making it more susceptible to possible errors.

To reduce or even eliminate the possible errors in generating control tables, various studies and articles on automatic generation and verification have been presented in the literature, using different formal tools like EURIS, Ladder logic, Petri Nets (PNs), RailML, Controlled Natural Language (CNL), Maple, B-method, Gröbner Bases (GBs), Abstract State Machines (ASMs), Finite State Machines (FSMs), etc. All the examined papers assumed that the topological layout of a specific railway station had to be prepared manually using dedicated input tools. However, substantial progress was achieved only when the topology of the specific railway station was generated in the corresponding graphical editor before generating the control table itself, which brings a specific level of automation to the whole process.

This paper will present a methodology for the generation of station control tables using the MATHEMATICA package directly from the AutoCAD station signaling layout. This is usually prepared as the first step in the design of a railway signaling system for specific railway stations.

Thin-walled structures are particularly advantageous for applications that require lightweight designs with high stiffness and strength. Therefore, understanding their mechanical behavior is essential. Internally stiffened thin-walled structures are of particular interest, as internal reinforcements can be designed to optimize the moment of inertia, leading to increased stiffness and strength. This investigation applies structural finite element analysis (FEA) to assess the strength behavior of internally reinforced hollow-box beams. A total of twelve different beams were subjected to static three-point bending stresses, with loads applied to beams that had previously undergone stiffness optimization for enhanced performance. These beams were carefully modified to achieve the highest possible stiffness while minimizing mass. The strength was evaluated using von Mises equivalent stress values, and various metrics were provided to analyze the behavior of the optimized models. The optimized beams were compared with both the initial models and a reference model of an unstiffened beam to assess the impact of stiffness-based optimization on strength results. The findings indicate that the optimization technique, originally developed to increase stiffness while reducing mass, also improves specific strength while maintaining mass reduction. The primary benefit of reducing material in certain parts is not only the decrease in material costs but also the enhanced motion capability of mobile components, which allows for faster movement. The technological complexity of producing such structures was high until recent years, when additive manufacturing emerged as an affordable and high-quality solution for fabricating complex parts like those studied here. Future research could focus on exploring the long-term stability and safety of these structures.

The purpose of this work is to develop and validate an analytical model that provides accurate predictions of the mechanical properties of hollow box beams, in comparison to traditional methods. The study focuses on numerical results obtained for a hollow box beam with a rectangular cross-section. To achieve this objective, a novel analytical model was created. A finite element model (FEM) of the box beam was built in the comercial FEM software ANSYS Mechanical ADPL, and the results were compared using classical theory, the new equation, and numerical techniques. Linear static analysis was performed to analyze the results, and a mathematical method was employed to compare the outcomes. Validation of the newly developed equation was performed by comparing it with both the numerical model and the classical equation, and this approach proved successful. It was shown that the new equation outperforms the classical equation in accurately predicting the mechanical behavior of the studied geometries. This superiority was demonstrated through error analysis, which revealed that the new equation resulted in lower errors than the classical equation. It was found that the maximum error between the analytical equation and the numerical method decreased from approximately 2.5% for the classical equation to around 0.24% using the derived equation.

The fabrication of conformal cooling channels (CCCs) has become increasingly efficient and cost-effective throughout the past few years. Conformal cooling channels, also known as CCCs, are superior to straight-drilled channels in terms of efficiency in injection molding engineering applications owing to their ability to provide superior cooling. The ability of CCCs to conform to the curves of molded geometry is the reason for this. The implementation of CCCs results in a significant reduction in the amount of time required for cooling, total injection time, thermal stresses, and warpage. Compared to the construction of a regular channel, the construction of a CCC is usually more complex. To maximize the development of designs that are both economically viable and highly efficient, computer-aided engineering (CAE) simulations are necessary. The purpose of this study is to conduct a sensitivity analysis of a full injection mold, with eight cooling channels, by means of a thermal analysis performed in 2D. The objective is to achieve the best possible placement of CCCs to diminish the amount of time required for ejection and to enhance the uniformity of temperature distribution. The results show that all the variables have significant sensitivity to the maximum and average temperature on the injected part and, therefore, are suitable for design optimization procedures.

Thin-walled structures are widely used for engineering applications where lightweight structures with high stiffness and high resistance are especially advantageous or even required. In this work, a novel analytical equation is developed to accurately predict the mechanical behavior of thin-walled beams. The Finite Element Method (FEM) was used to build the model and obtain the results. The newly developed equation is designed for calculating the displacement of a simply supported beam subjected to torsional loads, which are distributed at midspan using two triangular load functions applied in opposite directions in the FEM models. The Eureqa software was utilized to uncover hidden analytical models, which were subsequently validated. The goal is to provide a formula that allows for the comparison of analytical calculations with numerical results for combined bending and torsion loads. A FEM model of a hollow-box beam with a rectangular cross-section subjected to torsion was constructed, and analytical calculations were performed. The analytical results were compared with the numerical results to assess their accuracy, and good agreement was found. In the future, other models, such as internally reinforced beams, could be tested using this methodology. Additionally, different conditions could be applied to the model studied in this work to evaluate the limitations and validity of the developed analytical model.