AUIQ Technical Engineering Science

Wireless power transfer has attracted more and more attention as a potential solution to electric vehicle charging due to its safety, convenience and the potential to automate this process. This review paper gives an overview of inductive and capacitive wireless power transfer technologies, especially focusing on hybrid wireless power transfer systems that integrate both approaches. The basic operating principles in both inductive and capacitive power transfer are covered with an emphasis on their advantages and inherent limitations with respect to efficiency, voltage and current stress, the tolerance of misalignment and the power capability. Building upon this background, the paper is limited to hybrid systems as an effective mechanism for taking advantage of both the advantages of inductive and capacitive coupling and reducing the respective disadvantages. Special attention is paid to compensation networks used in HWPT systems. Different compensation topologies are reviewed and analyzed in terms of the role they play in improving the power transfer capability, improving operating conditions to the favor of the system and improving the system stability. In addition, various losses mechanisms related to the hybrid systems such as copper losses, magnetic losses, dielectric losses, and power electronic losses are analysed in order to check its effect on the overall system efficiency. This review is aimed to give a clear and structured understanding of HWPT systems and to bring out their potential as a possible solution for high-efficiency wireless charging applications, especially in the field of electric vehicle systems.

Additive manufacturing is increasingly being adopted in prosthetic manufacturing because digital workflows can reduce production time and improve repeatability compared to traditional lamination. This study measured two FDM polymers—polylactic acid (PLA) and polyethylene terephthalate glycol (PETG)—against a baseline of Perlon epoxy laminate (not including carbon reinforcement) using standardized coupon tests (n = 3 per test). PLA and PETG samples were printed at 100% fill with a 0.6 mm nozzle, while laminate coupons were prepared by wet layup and cut to the required geometry. Tensile testing followed ASTM D638 and three-point bending followed ASTM D790, modulus, 0.2% offset yield strength, and ultimate strength were derived from stress-strain curves. In tension, PETG showed higher UTS (38.71 ± 1.97 MPa), while PLA showed a slightly higher modulus (0.363 ± 0.104 GPa); The laminate showed lower tensile strength (28.33 ± 2.29 MPa). In bending, PETG achieved the highest bending strength (70.62 ± 3.15 MPa), while PLA showed higher bending stiffness (Ef = 2.07 ± 0.12 GPa) than PETG (1.72 ± 0.02 GPa). To strengthen the technical relevance, a simplified linear elastic finite element model of a three-point bending configuration was demonstrated, confirming the expected midpoint stress concentration and material-dependent deflection. Overall, the findings support evidence-based material screening for prosthetic socket construction, while socket-level validation under standardized structural loading is still needed.

Due to the generally unqualified nature of prediction data and the difficulty of interpreting predictions, predicting diabetes remains a significant hurdle in the adoption of machine learning within the medical domain. In this study, several tree-based machine learning techniques (LightGBM, XGBoost, CatBoost, and Gradient Boosting) were applied to predict diabetes using the 2015 BRFSS dataset, while two ensemble methods (soft voting and stacking) were employed to improve predictive accuracy. The performance analysis of the individual models and ensemble approaches indicates that CatBoost achieved the highest accuracy among the single classifiers (0.871), with an F1-score of 0.871 and a ROC–AUC of 0.921. Both ensemble methods further improved predictive performance compared with individual models. The soft voting ensemble obtained an overall accuracy of 0.878 and a ROC–AUC of 0.928, whereas the stacking ensemble achieved the highest overall performance, with an accuracy of 0.883, an F1-score of 0.883, and a ROC–AUC of 0.934. Moreover, SHAP-based analysis identified general health, body mass index, and age as the most influential factors affecting the prediction outcomes. We conclude that ensemble learning improves predictive performance while still providing a level of interpretability when assessing risk for developing diabetes.