Search Results (730)

Automated surface defect detection in steel manufacturing is pivotal for ensuring product quality, yet it remains an open challenge owing to the extreme heterogeneity of defect morphologies—ranging from hairline cracks and microscopic pores to elongated scratches and shallow dents. Existing approaches, whether classical vision pipelines or recent deep-learning paradigms, struggle to simultaneously satisfy the stringent demands of industrial scenarios: high accuracy on sub-millimeter flaws, insensitivity to texture-rich backgrounds, and real-time throughput on resource-constrained hardware. Although contemporary detectors have narrowed the gap, they still exhibit pronounced sensitivity–robustness trade-offs, particularly in the presence of scale-varying defects and cluttered surfaces. To address these limitations, we introduce MBY (MBDNet-Attention-YOLO), a lightweight yet powerful framework that synergistically couples the MBDNet backbone with the YOLO detection head. Specifically, the backbone embeds three novel components: (1) HGStem, a hierarchical stem block that enriches low-level representations while suppressing redundant activations; (2) Dynamic Align Fusion (DAF), an adaptive cross-scale fusion mechanism that dynamically re-weights feature contributions according to defect saliency; and (3) C2f-DWR, a depth-wise residual variant that progressively expands receptive fields without incurring prohibitive computational costs. Building upon this enriched feature hierarchy, the neck employs our proposed MultiSEAM module—a cascaded squeeze-and-excitation attention mechanism operating at multiple granularities—to harmonize fine-grained and semantic cues, thereby amplifying weak defect signals against complex textures. Finally, we integrate the Inner-SIoU loss, which refines the geometric alignment between predicted and ground-truth boxes by jointly optimizing center distance, aspect ratio consistency, and IoU overlap, leading to faster convergence and tighter localization. Extensive experiments on two publicly available steel-defect benchmarks—NEU-DET and PVEL-AD—demonstrate the superiority of MBY. Without bells and whistles, our model achieves 85.8% mAP@0.5 on NEU-DET and 75.9% mAP@0.5 on PVEL-AD, surpassing the best-reported results by significant margins while maintaining real-time inference on an NVIDIA Jetson Xavier. Ablation studies corroborate the complementary roles of each component, underscoring MBY’s robustness across defect scales and surface conditions. These results suggest that MBY strikes an appealing balance between accuracy, efficiency, and deployability, offering a pragmatic solution for next-generation industrial quality-control systems. Full article

In many developing African countries, milk safety is often managed through traditional methods such as fermentation or boiling over firewood. While these approaches reduce some microbial risks, they present critical limitations. Firewood dependency contributes to deforestation, depletion of agricultural residues, and loss of soil fertility, which, in turn, compromise environmental health and food security. Solar pasteurization provides a reliable and sustainable method for thermally inactivating pathogenic microorganisms in milk and other perishable foods at sub-boiling temperatures, preserving its nutritional quality. This study aimed to evaluate the thermal and microbial performance of a low-cost solar milk pasteurization system, hypothesized to effectively reduce microbial contaminants and retain milk quality under natural sunlight. The system was constructed using locally available materials and tailored to the climatic conditions of the Savanna ecological zone in West Africa. A flat-plate glass solar collector was integrated with a 0.15 cm thick stainless steel cylindrical milk vat, featuring a 2.2 cm hot water jacket and 0.5 cm thick aluminum foil insulation. The system was tested in Navrongo, Ghana, under ambient temperatures ranging from 30 °C to 43 °C. The pasteurizer successfully processed up to 8 L of milk per batch, achieving a maximum milk temperature of 74 °C by 14:00 GMT. Microbial analysis revealed a significant reduction in bacterial load, from 6.6 × 10⁶ CFU/mL to 1.0 × 10² CFU/mL, with complete elimination of coliforms. These results confirmed the device’s effectiveness in achieving safe pasteurization levels. The findings demonstrate that this locally built solar pasteurization system is a viable and cost-effective solution for improving milk safety in arid, electricity-limited regions. Its potential scalability also opens avenues for rural entrepreneurship in solar-powered food and water treatment technologies. Full article

In the present paper, a mathematical analysis of the Gardner equation with varying coefficients has been performed to give a more realistic model of physical phenomena, especially in regards to plasma physics. First, a Lie symmetry analysis was carried out, as a result of which a symmetry classification following the different representations of the variable coefficients was systematically derived. The reduced ordinary differential equation obtained is solved using the power-series method and solutions to the equation are represented graphically to give an idea of their dynamical behavior. Moreover, a fully connected neural network has been included as an efficient computation method to deal with the complexity of the reduced equation, by using traveling-wave transformation. The validity and correctness of the solutions provided by the neural networks have been rigorously tested and the solutions provided by the neural networks have been thoroughly compared with those generated by the Runge–Kutta method, which is a conventional and well-recognized numerical method. The impact of a variation in the loss function of different coefficients has also been discussed, and it has also been found that the dispersive coefficient affects the convergence rate of the loss contribution considerably compared to the other coefficients. The results of the current work can be used to improve knowledge on the nonlinear dynamics of waves in plasma physics. They also show how efficient it is to combine the approaches, which consists in the use of analytical and semi-analytical methods and methods based on neural networks, to solve nonlinear differential equations with variable coefficients of a complex nature. Full article

This study analyzes the impact of geometric variations induced by the manufacturing process on the aerodynamic efficiency of an airfoil used in the design of a 3 kW wind turbine blade. For this purpose, a computational fluid dynamics (CFD) analysis was implemented, and the results were compared with those obtained using QBlade software. After blade fabrication, experimental evaluation was performed using the laser triangulation technique, enabling the reconstruction of the deformed airfoils and their comparison with the original geometry. Additional CFD simulations were carried out on the manufactured airfoil to quantify the loss of aerodynamic efficiency due to geometrical deformations. The results show that the geometric deviations significantly affect the aerodynamic coefficients, generating a decrease in the lift coefficient and an increase in the drag coefficient, which negatively impacts the airfoil aerodynamic efficiency. A 14.9% reduction in the rotor power coefficient was observed with the deformed airfoils compared to the original design. This study emphasizes the importance of quality control in wind turbine blade manufacturing processes and its impact on turbine power performance. In addition, the findings can contribute to the development of design compensation strategies to mitigate the adverse effects of geometric imperfections on the aerodynamic performance of wind turbines. Full article

AlGaN-based high-electron-mobility transistors are critical for next-generation power electronics and radio-frequency applications, yet achieving stable enhancement-mode operation with a high threshold voltage remains a key challenge. In this work, we designed p-AlGaN superlattices with different structures and performed energy band structure simulations using the device simulation software Silvaco. The results demonstrate that thin barrier structures lead to reduced acceptor incorporation, thereby decreasing the number of ionized acceptors, while facilitating vertical hole transport. Superlattice samples with varying periodic thicknesses were grown via metal-organic chemical vapor deposition, and their crystalline quality and electrical properties were characterized. The findings reveal that although gradient-thickness barriers contribute to enhancing hole concentration, the presence of thick barrier layers restricts hole tunneling and induces stronger scattering, ultimately increasing resistivity. In addition, we simulated the structure of the enhancement-mode HEMT with p-AlGaN as the under-gate material. Analysis of its energy band structure and channel carrier concentration indicates that adopting p-AlGaN superlattices as the under-gate material facilitates achieving a higher threshold voltage in enhancement-mode HEMT devices, which is crucial for improving device reliability and reducing power loss in practical applications such as electric vehicles. Full article

The efficiency decline of Francis turbines, a key component of hydroelectric power generation, presents a multifaceted challenge influenced by interconnected factors such as water quality, incidence angle, erosion, and runner wear. This paper is structured into two main sections to address these issues. The first section applies the Buckingham π theorem to establish a dimensional analysis (DA) framework, providing insights into the relationships among the operational variables and their impact on turbine wear and efficiency loss. Dimensional analysis offers a theoretical basis for understanding the relationships among operational variables and efficiency within the scope of this study. This understanding, in turn, informs the selection and interpretation of features for machine learning (ML) models aimed at the predictive maintenance of the target variable and important features for the next stage. The second section analyzes an extensive dataset collected from a Francis turbine in Colombia, a country that is heavily reliant on hydroelectric power. The dataset consisted of 60,501 samples recorded over 15 days, offering a robust basis for assessing turbine behavior under real-world operating conditions. An exploratory data analysis (EDA) was conducted by integrating linear regression and a time-series analysis to investigate efficiency dynamics. Key variables, including power output, water flow rate, and operational time, were extracted and analyzed to identify patterns and correlations affecting turbine performance. This study seeks to develop a comprehensive understanding of the factors driving Francis turbine efficiency loss and to propose strategies for mitigating wear-induced performance degradation. The synergy lies in DA’s ability to reduce dimensionality and identify meaningful features, which enhances the ML models’ interpretability, while ML leverages these features to model non-linear and time-dependent patterns that DA alone cannot address. This integrated approach results in a linear regression model with a performance (R²-Test = 0.994) and a time series using ARIMA with a performance (R²-Test = 0.999) that allows for the identification of better generalization, demonstrating the power of combining physical principles with advanced data analysis. The preliminary findings provide valuable insights into the dynamic interplay of operational parameters, contributing to the optimization of turbine operation, efficiency enhancement, and lifespan extension. Ultimately, this study supports the sustainability and economic viability of hydroelectric power generation by advancing tools for predictive maintenance and performance optimization. Full article

Distributed energy resource systems offer improved energy utilization and reduced transmission losses by decentralizing power generation and load management. However, the power quality is often compromised by inefficient customer-side equipment, such as single-phase induction motors, which suffer from low efficiency and poor power factor. To address this issue, this paper proposes a permanent magnet retrofitting method for single-phase induction motors, which replaces the squirrel-cage rotor with a permanent magnet rotor while preserving the original stator and winding structure. The proposed method aims to enhance motor performance without significant structural changes. A single-phase induction motor was retrofitted using the proposed method, and its performance was evaluated through finite element simulations to verify the effectiveness of the design approach. This study also investigated the key factors influencing motor starting performance after the introduction of permanent magnets. This study presents a practical and effective method for the permanent magnet retrofitting of single-phase induction motors, which contributes to improving motor efficiency and enhancing power quality in distributed energy resource systems. Full article

The increasing penetration of renewable energy sources (RESs), particularly solar photovoltaic (PV) sources, has introduced significant uncertainty into power system operations, challenging traditional scheduling models and threatening system reliability. This study proposes a Fuzzy Unit Commitment Model (FUCM) designed to address uncertainty in load demand, solar PV generation, and spinning reserve requirements by applying fuzzy linear programming techniques. The FUCM reformulates uncertain constraints using triangular membership functions and integrates them into a mixed-integer linear programming (MILP) framework. The model’s effectiveness is demonstrated through two case studies: a 30-generator test system and a national-scale power system in Thailand comprising 171 generators across five service zones. Simulation results indicate that the FUCM consistently produces stable scheduling solutions that fall within deterministic upper and lower bounds. The model improves reliability metrics, including reduced loss-of-load probability and minimized load deficiency, while maintaining acceptable computational performance. These results suggest that the proposed approach offers a practical and scalable method for unit commitment planning under uncertainty. By enhancing both operational stability and economic efficiency, the FUCM contributes to the sustainable management of RES-integrated power systems. Full article

The increase in power generation facilities from nonprogrammable renewable sources is posing several challenges for the management of electrical systems, due to phenomena such as congestion and reverse power flows. In mitigating these phenomena, Power-to-Gas plants can make an important contribution. In this paper, a linear optimisation study is presented for the sizing of a Power-to-Hydrogen plant consisting of a PEM electrolyser, a hydrogen storage system composed of multiple compressed hydrogen tanks, and a fuel cell for the eventual reconversion of hydrogen to electricity. The plant was sized with the objective of minimising reverse power flows in a medium-voltage distribution network characterised by a high presence of photovoltaic systems, considering economic aspects such as investment costs and the revenue obtainable from the sale of hydrogen and excess energy generated by the photovoltaic systems. The study also assessed the impact that the electrolysis plant has on the power grid in terms of power losses. The results obtained showed that by installing a 737 kW electrolyser, the annual reverse power flows are reduced by 81.61%, while also reducing losses in the transformer and feeders supplying the ring network in question by 17.32% and 29.25%, respectively, on the day with the highest reverse power flows. Full article

Large turbine generators rely on shaft earthing brushes to safely divert harmful shaft currents to ground, protecting bearings from electrical damage. This paper presents a novel deep learning-based diagnostic framework to detect and classify faults in shaft earthing brushes of large turbine generators. A key innovation lies in the use of FFT-derived spectrograms from both voltage and current waveforms as dual-channel inputs to the CNN, enabling automatic feature extraction of time–frequency patterns associated with different SEB fault types. The proposed framework combines advanced signal processing and convolutional neural networks (CNNs) to automatically recognize fault-related patterns in shaft grounding current and voltage signals. In the approach, raw time-domain signals are converted into informative time–frequency representations, which serve as input to a CNN model trained to distinguish normal and faulty conditions. The framework was evaluated using data from a fleet of large-scale generators under various brush fault scenarios (e.g., increased brush contact resistance, loss of brush contact, worn out brushes, and brush contamination). Experimental results demonstrate high fault detection accuracy (exceeding 98%) and the reliable identification of different fault types, outperforming conventional threshold-based monitoring techniques. The proposed deep learning framework offers a novel intelligent monitoring solution for predictive maintenance of turbine generators. The contributions include the following: (1) the development of a specialized deep learning model for shaft earthing brush fault diagnosis, (2) a systematic methodology for feature extraction from shaft current signals, and (3) the validation of the framework on real-world fault data. This work enables the early detection of brush degradation, thereby reducing unplanned downtime and maintenance costs in power generation facilities. Full article

With the development and application of electric vehicles powered by lithium-ion batteries (LIBs) at high altitude, the lack of research on the aging laws and mechanisms of LIBs under a low-pressure aviation environment has become an important obstacle to their safe application. Herein, the influences and mechanisms of high-altitude and low-pressure environment (50 kPa) on the cycling performance of commercial pouch LIBs were systematically studied. The results showed that low air pressure caused a sharp decrease in battery capacity to 46.6% after 200 cycles, with a significant increase in charge transfer impedance by 70%, and the contribution rate of active lithium loss reached 74%. Low air pressure led to irreversible deformation of the battery, resulting in the expansion of the gap between the electrodes, poor electrolyte infiltration, and reduction of the effective lithium insertion area, which in turn induced multiple synergistic accelerated decay mechanisms, including obstructed lithium-ion transmission, reduced interfacial reaction efficiency, increased active lithium consumption, changes in heat generation structure, and a significant increase in heat generation. After applying external force, the deformation of the electrode was effectively suppressed, and the cycle capacity retention rate increased to 87.6%, which significantly alleviated the performance degradation of LIBs in low pressure environment. This work provides a key theoretical basis and engineering solutions for the design of power batteries in high-altitude areas. Full article

In the current context of environmental concerns and the search for sustainable food solutions, this study investigated the valorization of Luffa cylindrica seed press cake, a waste byproduct from oil extraction, as a functional ingredient for meat substitute formulations. The research systematically characterized the functional and bioactive properties of L. cylindrica seed press cake powder (LP) and its blends with tapioca flour (TF) at ratios of 30–70%. Techno-functional analyses included: hydration properties (water holding capacity, water absorption capacity, water absorption index, water solubility index, swelling power, oil absorption capacity); rheological characteristics; bioactive profiling through antioxidant assays (DPPH, ABTS, FRAP); and reducing sugar content determination. Meat substitute formulations were developed using an LP30/TF70 blend combined with coral lentils, red beet powder, and water, followed by a sensory evaluation and storage stability assessment. Pure L. cylindrica powder exhibited the highest water holding capacity (3.62 g H₂O/g) and reducing sugar content (8.05 mg GE/g), while tapioca flour showed superior swelling properties. The blends demonstrated complementary functional characteristics, with the LP30/TF70 formulation selected for meat substitute development based on optimal textural properties. The sensory evaluation revealed significant gender differences in acceptance, with women rating the product substantially higher than men across all attributes. The study successfully demonstrated the feasibility of transforming agricultural waste into a valuable functional ingredient, contributing to sustainable food production and representing the first comprehensive evaluation of L. cylindrica seed press cake for food applications. However, the study revealed limitations, including significant antioxidant loss during thermal processing (80–85% reduction); a preliminary sensory evaluation with limited participants showing gender-dependent acceptance; and a reliance on locally available tapioca flour, which may limit global applicability. Future research should focus on processing optimization to preserve bioactive compounds, comprehensive sensory studies with diverse populations, and an investigation of alternative starch sources to enhance the worldwide implementation of this valorization approach. Full article

Background: Limited availability of high-quality labeled biomedical image datasets presents a significant challenge for training deep learning models in medical diagnostics. This study proposes a novel image generation framework combining conditional generative adversarial networks (cGANs) with a Mixture-of-Experts (MoE) architecture and color histogram-aware loss functions to enhance synthetic blood cell image quality. Methods: RGB microscopic images from the BloodMNIST dataset (eight blood cell types, resolution 3 × 128 × 128) underwent preprocessing with k-means clustering to extract the dominant colors and UMAP for visualizing class similarity. Spearman correlation-based distance matrices were used to evaluate the discriminative power of each RGB channel. A MoE–cGAN architecture was developed with residual blocks and LeakyReLU activations. Expert generators were conditioned on cell type, and the generator’s loss was augmented with a Wasserstein distance-based term comparing red and green channel histograms, which were found most relevant for class separation. Results: The red and green channels contributed most to class discrimination; the blue channel had minimal impact. The proposed model achieved 0.97 classification accuracy on generated images (ResNet50), with 0.96 precision, 0.97 recall, and a 0.96 F1-score. The best Fréchet Inception Distance (FID) was 52.1. Misclassifications occurred mainly among visually similar cell types. Conclusions: Integrating histogram alignment into the MoE–cGAN training significantly improves the realism and class-specific variability of synthetic images, supporting robust model development under data scarcity in hematological imaging. Full article

Transformer-based models have demonstrated outstanding performance in trajectory prediction; however, their complex architecture demands substantial computing power, and their performance degrades significantly in long-term prediction. A transformer model was developed to predict vehicle trajectory in urban low-speed T-intersections. Microscopic traffic simulation data were generated to train the trajectory-prediction model; furthermore, validation data focusing on atypical scenarios were also produced. The appropriate loss function to improve prediction accuracy was explored, and the optimal input/output sequence length for efficient data management was examined. Various driving-characteristics data were employed to evaluate the model’s generalization performance. Consequently, the smooth L1 loss function showed outstanding performance. The optimal length for the input and output sequences was found to be 1 and 3 s, respectively, for trajectory prediction. Additionally, improving the model structure—rather than diversifying the training data—is necessary to enhance generalization performance in atypical driving situations. Finally, this study confirmed that the additional features such as vehicle position and speed variation extracted from the original trajectory data decreased the model accuracy by about 21%. These findings contribute to the development of applicable lightweight models in edge computing infrastructure to be installed at intersections, as well as the development of a trajectory prediction and accident analysis system for various scenarios. Full article

Laser-based powder bed fusion of metals (PBF-LB/M) is one of the most promising additive manufacturing technologies to fabricate complex-structured metal parts. However, its corresponding applications have been limited by technical bottlenecks and increasingly strict industrial requirements. Process optimization, a scientific issue, urgently needs to be solved. In this paper, a three-phase transient model based on the level-set method is established to examine the heat transfer and melt pool behavior in PBF-LB/M. Surface tension, the Marangoni effect, and recoil pressure are implemented in the model, and evaporation-induced mass and thermal loss are fully considered in the computing element. The results show that the surface roughness and density of metal parts induced by heat transfer and melt pool behavior are closely related to process parameters such as laser power, layer thickness, scanning speed, etc. When the volumetric energy density is low, the insufficient fusion of metal particles leads to pore defects. When the line energy density is high, the melt track is smooth with low porosity, resulting in the high density of the products. Additionally, the partial melting of powder particles at the beginning and end of the melting track usually contributes to pore formation. These findings provide valuable insights for improving the quality and reliability of metal additive manufacturing. Full article