Ensuring mission reliability is vital for the autonomous deployment of unmanned aerial vehicles (UAVs) in modern power and energy systems, particularly under spatial and operational constraints. This study presents a data-driven classification method that assesses the reliability of UAV-based inspection missions by identifying whether individual mission locations are suitable, at risk, or infeasible based on spatial and operational parameters. Leveraging the Cumulative UAV Routing Problem (CUAVRP) benchmark, four representative mission scenarios were analyzed, each characterized by unique UAV fleet sizes, sensor ranges, and endurance limits. Synthetic stress nodes were introduced to emulate edge-case conditions encountered in infrastructure inspection tasks. Each node was classified based on three categorical targets: Mission Feasibility, Coverage Reliability, and Deployment Suitability. A gradient boosting classification model was trained on spatial and operational features to determine node status. Evaluation across all scenarios yielded consistently high performance, with the cuavrp_d9_k6_r800 scenario achieving 97.05% accuracy, 96.33% precision, 97.72% recall, and 97.02% F1-score. Furthermore, incorporating physical-layer degradation factors such as signal attenuation, multipath fading, and interference is expected to enhance the realism of future reliability assessments and improve classification robustness. The proposed classification framework supports intelligent mission planning, enhances operational resilience, and facilitates automated UAV deployment strategies in critical inspection environments within the power and energy sector.

The MIMO antenna design is specifically engineered to support optimized performance in emerging 6G networks. Utilizing advanced techniques such as metamaterials and machine learning algorithms, the antenna system achieves high data rates, improved diversity, and robust signal reliability, making it ideal for next-generation ultra-fast and intelligent wireless communication technologies. Our advanced metamaterial configuration demonstrates high gain and bandwidth. A low ECC value 0.0004 shows minimal correlation, ensuring better signal diversity and improved system performance. Similarly, a high diversity gain confirms the antenna’s efficiency in maintaining robust signal reception under varying conditions. The CCL values of 0.0916 bits/Hz bits/Hz provide insight into the information-carrying capacity of the MIMO configuration. The MIMO antenna design achieves a maximum gain of 8.9 dBi and a wide bandwidth of 30 THz. This performance is attained through a combination of parametric optimization and machine learning techniques, enhancing both efficiency and operational range. The machine learning algorithms used for optimization yield a high R² value of 0.99, indicating excellent prediction accuracy. The proposed antenna, featuring metamaterial characteristics, demonstrates strong potential for next-generation 6G networks, offering enhanced performance, efficiency, and compact design integration.

Blood cancer can be fatal if not detected early; innovative biosensors with machine learning optimization enable timely diagnosis by identifying cancer-specific biomarkers in blood, improving survival rates through earlier intervention and targeted treatment. A graphene-based sensor, crafted with advanced materials, enhances sensitivity for rapid and early blood cancer detection, offering improved diagnostic accuracy and timely medical intervention for better patient outcomes. Machine learning optimization is used to achieve higher sensitivity. The graphene sensor achieves a maximum sensitivity of 1430 nm/RIU, enabling highly accurate and efficient blood cancer detection performance. The developed sensor demonstrates an impressive detection limit of 0.044, offering exceptional precision and sensitivity, making it highly effective for early-stage blood cancer diagnosis and clinical applications. The optimized sensor design achieves a high-quality factor of 125 and a figure of merit of 121, indicating excellent performance, sharp resonance, and enhanced precision for blood cancer detection applications. Optimization is achieved using parametric optimization. Optimization of the sensor is accomplished through detailed parametric analysis, resulting in a finely tuned design. This optimized structure significantly enhances sensitivity and detection speed, making it a highly suitable choice for early-stage and rapid blood cancer diagnosis, thereby improving the chances of timely treatment and patient survival.

Accurate assessment of transformer insulating paper is vital for reliable operation and optimal transformer management, with the Degree of Polymerization (DP) serving as a primary indicator of insulation health. Direct DP measurement is often impractical, prompting this study to explore machine learning models for predicting DP using 2-Furfuraldehyde (2-FAL), a cellulose degradation byproduct measurable in transformer oil. This approach classifies insulation into four categories—Fresh (DP: 700–1200), Lightly Aged (DP: 450–700), Moderately Aged (DP: 250–450), and Worstly Aged (DP < 250)—based on DP values, offering a streamlined alternative to conventional multi-gas diagnostic methods. Supervised machine learning algorithms were developed using IEEE C57.104-2019 standard data, employing regression (Linear Regression, Polynomial Regression, Random Forest Regressor) to predict continuous DP and classification (Logistic Regression, Support Vector Machine with RBF kernel, Random Forest Classifier) to categorize insulation condition. Model performance was evaluated using regression metrics (Mean Squared Error, Mean Absolute Error, R² Score) and classification metrics (accuracy, precision, recall, F1-score). The Random Forest Regressor (R²: 0.894) and Classifier (accuracy: 0.925) demonstrated superior performance, enabling precise, non-invasive DP estimation and condition assessment. These findings highlight the efficacy of 2-FAL-based machine learning models for transformer health monitoring, facilitating predictive maintenance and enhancing operational reliability.

The accurate prediction of compressible high-speed jet flows remains a critical challenge in computational aerodynamics due to the strong coupling between shock waves, expansion fans, shear-layer turbulence, and mixing processes downstream of nozzles. While density-based (DB) solvers are conventionally used for supersonic and transonic regimes, pressure-based (PB) solvers have recently gained attention for their reduced computational cost; however, their suitability for underexpanded jet modeling remains insufficiently explored. This study provides a systematic evaluation of PB and DB solvers available in ANSYS Fluent for simulating compressible flow discharged from an axisymmetric convergent nozzle over a range of nozzle pressure ratios (PR = 1.92–5). Experimental Schlieren flow visualization was conducted using a custom optical setup to qualitatively assess shock structures and validate near-field flow features. Numerical simulations employed compressible Reynolds Average Navier-Stokes (RANS) formulations with a standard k–ε turbulence model, structured quadrilateral meshes, and consistent boundary conditions for both solvers. Validation against published measurements demonstrated that both solvers accurately predict Mach disk formation and the streamwise location of the first shock cell, with a maximum deviation of 4.5% in peak Mach number. For PR > 1.92, both solvers captured the characteristic diamond shock pattern and the progressive increase in shock cell strength and spacing; the first shock cell occurred at X/De ≈ 1.09, 1.57, and 2 for PR = 3, 4, and 5, respectively. While PB and DB solvers exhibited comparable performance in resolving centerline Mach number and pressure oscillations, the DB solver overpredicted turbulent kinetic energy in the far-field subsonic region due to its known sensitivity at low Mach numbers. Discharge coefficients predicted by both solvers showed close agreement, with differences < 0.2%. Results demonstrate that the PB solver, despite being traditionally associated with incompressible and low-speed flows, can reliably model underexpanded supersonic jets at significantly reduced numerical cost. The findings provide practical guidance for CFD practitioners seeking cost-effective tools for compressible nozzle flow modeling and contribute to broader discussions on solver strategy selection for high-speed aerodynamic simulations.

The growing demand for high-speed communication necessitates antennas operating at higher frequencies in the THz range with broad bandwidth, compact size and higher gain. In this research, we have proposed a compact THz antenna designed with metamaterial-inspired elements to achieve improved gain and wideband performance. This paper presents the design and performance analysis of a high-efficiency MIMO antenna system for terahertz (THz) communication applications. The proposed antenna is optimized to deliver robust diversity performance, low mutual coupling, and minimal channel degradation. The design is also optimized with machine learning, with the highest R ² value of 0.95. The optimized design gives 54 THz bandwidth and 7.6 dBi gain. The ECC remains well below 0.005, ensuring excellent isolation between elements. The TARC exhibits values below − 10 dB across the critical bandwidth, confirming low reflection under simultaneous port excitation. These results demonstrate the designed structure’s potential for integration into THz communication and THz Wireless Personal Area Network.