Search Results (73,899)

Electric aviation is the future development direction of aviation industry technology. Electric vertical take-off and landing aircraft(eVTOL) is an important carrier of electric aviation, whose technology research and development, processing and manufacturing, airworthiness certification and industrialization boom have been set off around the world. The electric propulsion technology has achieved rapid development as the key technology of eVTOL. Aiming at the demand for high torque density and high reliability of electric propulsion system, the paper analyzed the technical indexes of electric motor products of domestic and foreign benchmark enterprises. The key technologies such as motor integration, new electromagnetic topology, lightweight structure design, and high efficiency cooling is studied. It is pointed out that in order to pursue the high torque density and fault-tolerance performance, the integrated precise modeling of motor and controller, advanced materials and manufacturing technology are the development trend of the electric propulsion technology. The breakthrough of eVTOL electric propulsion technology can accelerate the commercial operation of civil eVTOL and promote the development of new quality productive forces. Full article

In rapidly urbanizing regions, encroachment on native green spaces has exacerbated ecological issues such as urban heat islands and flooding. Accurate mapping of tree species distribution is therefore vital for sustainable urban management. However, the high heterogeneity of urban landscapes, resulting from the coexistence of diverse land covers, built infrastructure, and anthropogenic activities, often leads to reduced robustness and transferability of remote sensing classification methods across different images and regions. In this study, we used very high–resolution Pléiades imagery and field-verified samples of eight common urban trees and background land covers. By employing transfer learning with advanced segmentation networks, we evaluated each model’s accuracy, robustness, and efficiency. The best-performing network delivered markedly superior classification consistency and required substantially less training time than a model trained from scratch. These findings offer concise, practical guidance for selecting and deploying deep learning methods in urban tree species mapping, supporting improved ecological monitoring and planning. Full article

This study employs RGB imagery and machine learning techniques to detect Tagosodes orizicolus infestations in “Tinajones” rice crops during the flowering stage, a critical challenge for agriculture in northern Peru. High-resolution images were acquired using an unmanned aerial vehicle (UAV) and preprocessed by extracting 256 × 256-pixel segments, focusing on three classes: infested zones, non-cultivated areas, and healthy rice crops. A dataset of 1500 images was constructed and utilized to train deep learning models based on VGG16 and ResNet50. Both models exhibited highly comparable performance, with VGG16 attaining a precision of 98.274% and ResNet50 achieving a precision of 98.245%, demonstrating their effectiveness in identifying infestation patterns with high reliability. To automate the analysis of complete UAV-acquired images, a web-based application was developed. This system receives an image, segments it into grids, and preprocesses each section using resizing, normalization, and dimensional adjustments. The pretrained VGG16 model subsequently classifies each segment into one of three categories: infested zone, non-cultivated area, or healthy crop, overlaying the classification results onto the original image to generate an annotated visualization of detected areas. This research contributes to precision agriculture by providing an efficient and scalable computational tool for early infestation detection, thereby supporting timely intervention strategies to mitigate potential crop losses. Full article

Construction and Demolition Waste (CDW) forecasting is essential for sustainable waste management and circular economy objectives. Traditional prediction models often face limitations when dealing with small datasets and extreme variability. This study introduces a robust statistical framework that employs the median and Median Absolute Deviation (MAD), applied to standardized CDW indicators: tons per day (t day⁻¹) and tons per square meter (t m⁻²). The method enables the detection of statistical anomalies using a ±2·MAD threshold, increasing the model’s resilience to outliers and enhancing its predictive reliability. The analysis is based on a dataset of 16 construction and rehabilitation projects, carried out under consistent technical methodologies, operational practices, and centralized data collection protocols. The results show that median-based predictions offer greater stability than mean-based estimators, particularly in skewed datasets. The framework successfully identifies projects with significant deviations, supporting targeted audits, performance monitoring, and iterative model refinement. Although the current model focuses on the duration and area as predictors, future enhancements should incorporate additional contextual variables. Furthermore, embedding the median–MAD logic within machine learning architectures (e.g., LSTM, ARIMAX) could improve scalability and support real-time CDW monitoring. These findings contribute to the development of data-driven forecasting tools that are aligned with operational efficiency and circularity goals in the construction sector. Full article

Volatile organic compounds (VOCs) in urine headspace are potential biomarkers for different medical conditions, as canines can detect human diseases simply by smelling VOCs. Because dogs can detect disease-specific VOCs, gas chromatography–mass spectrometry (GC–MS) systems may be able to differentiate medical conditions with enhanced accuracy and precision, given they have unprecedented efficiency in separating, quantifying, and identifying VOCs in urine. Advancements in instrumentation have permitted the development of portable GC–MS systems that analyze VOCs at the point of care, but these are designed for environmental monitoring, emergency response, and manufacturing/processing. The purpose of this study is to repurpose the HAPSITE^® ER portable GC–MS for identifying urinary VOC biomarkers. Method development focused on optimizing sample preparation, off-column conditions, and instrumental parameters that may affect performance. Once standardized, the method was used to analyze a urine standard (n = 10) to characterize intra-day reproducibility. To characterize inter-day performance, n = 3 samples each from three volunteers (and the standard) were analyzed each day for a total of four days (n = 48 samples). Results showed the method could detect VOC signals with adequate reproducibility and distinguish VOC profiles from different volunteers with 100% accuracy. Full article

Road transport systems (RTS) play an essential role in global supply chains, facilitating the efficient transport of goods and services over long distances and thus supporting economic activity on a worldwide scale. However, these systems face numerous challenges, particularly regarding safety, cost, and route optimization, requiring innovative and practical solutions to improve their overall performance. This paper proposes an in-depth analysis of RTS features forming a detailed dataset collected on the route between Dakhla (Morocco) and Paris (France). The study relies on applying advanced mathematical modeling techniques and analyzing several datasets to train various machine learning algorithms. The main objective is to identify optimized routes, combining high safety standards, reduced costs, and shorter transport times. The results show that the adopted approach results in safer and more efficient routes and complies with operational and regulatory constraints. Furthermore, this analysis highlights the importance of data quality and the integration of advanced technologies to deliver an intelligent route optimization system with significant reductions in cost and time. Finally, our results reveal that neural networks outperform other algorithms in this field, proving their superior effectiveness for this specific application. Full article

During the drilling process, stick–slip vibrations are one of the critical causes of bottom-hole assembly (BHA) failure and reduced drilling efficiency. To address this, this study first proposes a drill-string model based on a three-dimensional nonlinear finite beam element, combined with Hamilton’s principle of virtual work, to comprehensively describe the nonlinear behavior of the drill-string system. Next, to improve computational efficiency, the model is reduced using the modal truncation method, which retains the key modes of drill-string vibrations. Based on this, a model predictive control (MPC) method is designed to eliminate stick–slip vibrations. Furthermore, the robustness of the MPC method under parameter uncertainties is also investigated. In particular, the impact of the weight on bit (WOB) on the drill bit’s torsional velocity is further considered, and an MPC angular velocity comprehensive control scheme based on the dynamic WOB (DWOB-MPC) is proposed. This scheme stabilizes the velocity of the drill bit by dynamically adjusting the WOB, thereby eliminating stick–slip vibrations. Simulation results demonstrate that both the proposed MPC and DWOB-MPC methods effectively suppress stick–slip vibrations. Notably, the DWOB-MPC method further reduces the settling time and overshoot, exhibiting superior dynamic performance. Full article

Drug carriers hold significant promise for precision medicine but face persistent challenges in balancing high encapsulation efficiency with structural preservation during active loading. In this study, we present a microelectromechanical system (MEMS)-driven platform that can generate gigahertz (GHz)-frequency acoustic streaming (1.55 GHz) to enable nondestructive, power-tunable drug encapsulation in lipid vesicles. Utilizing DSPE-PEG-modified bilayers with hydrodynamic shear forces, our method achieves transient membrane permeability that preserves membrane integrity while permitting controlled doxorubicin (DOX) influx. We developed the GHz acoustic MEMS platform and applied it to systematically investigate two drug loading strategies: (1) loading DOX into giant unilamellar vesicles (GUVs, >10 μm in diameter) prior to extrusion into small unilamellar vesicles (SUVs, 100 nm) versus (2) direct acoustic loading into pre-formed SUVs. The GUV-first approach demonstrated better performance, achieving 60.04% ± 1.55% encapsulation efficiency (EE%) at 250 mW acoustic power—a 5.93% enhancement over direct SUV loading (54.11% ± 0.72%). Structural analysis via TEM confirmed intact SUV morphology post-loading, while power-dependent EE% analysis showed a linear trend. This work bridges gaps in nanocarrier engineering by optimizing drug loading strategies, aiming to offer a potential drug carrier platform for drug delivery in biomedical treatment in future. Full article

Orthopedics is undergoing a transformative shift driven by personalized medical technologies that enhance precision, efficiency, and patient outcomes. Virtual surgical planning, robotic assistance, and real-time 3D navigation have revolutionized procedures like total knee arthroplasty and hip replacement, offering unparalleled accuracy and reducing recovery times. Integrating artificial intelligence, advanced imaging, and 3D-printed patient-specific implants further elevates surgical precision, minimizes intraoperative complications, and supports individualized care. In sports orthopedics, wearable sensors and motion analysis technologies are revolutionizing diagnostics, injury prevention, and rehabilitation, enabling real-time decision-making and improved patient safety. Health-tracking devices are advancing recovery and supporting preventative care, transforming athletic performance management. Concurrently, breakthroughs in biologics, biomaterials, and bioprinting are reshaping treatments for cartilage defects, ligament injuries, osteoporosis, and meniscal damage. These innovations are poised to establish new benchmarks for regenerative medicine in orthopedics. By combining cutting-edge technologies with interdisciplinary collaboration, the field is redefining surgical standards, optimizing patient care, and paving the way for a highly personalized and efficient future. Full article

Accurate water body detection is essential for autonomous navigation and operational planning of unmanned surface vehicles (USVs). To address model adaptability to ambiguous boundaries caused by diverse scenarios and climatic conditions, this study proposes GFANet (Global–Local Feature Attention Network) for the real-time water surface semantic segmentation of camera-captured images. First, a Global–Local Feature (GLF) extraction module is proposed, integrating a self-attention-based local feature extractor and a multi-scale global feature extractor for parallel feature learning, thereby enhancing hierarchical feature representation. Second, a Gated Attention (GA) module is designed with a dual-branch gating mechanism to implement noise suppression and efficient low-level feature utilization. The method was validated on three publicly available datasets in relevant domains. The experimental results on the Riwa dataset show that GFANet achieves state-of-the-art segmentation performance (4.41 M parameters, 7.15 GFLOPs) with an mIoU of 82.29% and an mPA of 89.49%. Comparable performance metrics were obtained on the USVInland and WaterSeg datasets. Additionally, GFANet achieves a 154.98 FPS processing speed, meeting real-time segmentation requirements. The experimental results verify that GFANet achieves an optimal balance between high segmentation accuracy and real-time processing efficiency. Full article

Explainable Artificial Intelligence (XAI) is a well-established and dynamic field defined by an active research community that has developed numerous effective methods for explaining and interpreting the predictions of advanced machine learning models, including deep neural networks. Clustered Federated Learning (CFL) mitigates the difficulties posed by heterogeneous clients in traditional federated learning by categorizing related clients according to data characteristics, facilitating more tailored model updates, and improving overall learning efficiency. This paper introduces Explainable Clustered Federated Learning (XCFL), which adds explainability to clustered federated learning. Our method improves performance and explainability by selecting features, clustering clients, training local clients, and analyzing contributions using SHAP values. By incorporating feature-level contributions into cluster and global aggregation, XCFL ensures a more transparent and data-driven model update process. Weighted aggregation by feature contributions improves consumer diversity and decision transparency. Our results show that XCFL outperforms FedAvg and other clustering methods. Our feature-based explainability strategy improves model performance and explains how features affect clustering and model adjustments. XCFL’s improved accuracy and explainability make it a promising solution for heterogeneous and distributed learning environments. Full article

Lithium-ion batteries are extensively utilized in modern applications due to their high energy density, long cycle life, and efficiency. With the increasing demand for sustainable energy storage solutions, accurately estimating the State of Health (SOH) is essential to address challenges related to battery degradation and secondary life management. Electrochemical Impedance Spectroscopy (EIS) is a widely used diagnostic tool for evaluating battery performance due to its simplicity and cost-effectiveness. However, EIS often struggles to decouple overlapping electrochemical processes. The Distribution of Relaxation Times (DRT) method has emerged as a powerful alternative, enabling the isolation of key processes, such as ohmic resistance, SEI resistance, charge transfer resistance, and diffusion, thereby providing deeper insights into battery aging mechanisms. This paper presents a novel approach for estimating the State of Health (SOH) of batteries by leveraging DRT parameters across multiple State of Charge (SOC) levels. This study incorporates data from three lithium-ion batteries, each with distinct initial capacities, introducing variability that reflects the natural differences observed in real-world battery performance. By employing a Long Short-Term Memory (LSTM)-based machine learning model, the proposed framework demonstrates a superior accuracy in SOH prediction compared to traditional EIS-based methods. The results highlight the sensitivity of DRT parameters to SOH degradation and validate their effectiveness as reliable indicators for battery health. This research underscores the potential of combining a DRT analysis with AI-driven models to advance scalable, precise, and interpretable battery diagnostics. Full article

In response to the decreased accuracy in person detection caused by densely populated areas and mutual occlusions in public spaces, a human head-detection approach is employed to assist in detecting individuals. To address key issues in dense scenes—such as poor feature extraction, rough label assignment, and inefficient pooling—we improved the YOLOv7 network in three aspects: adding attention mechanisms, enhancing the receptive field, and applying multi-scale feature fusion. First, a large amount of surveillance video data from crowded public spaces was collected to compile a head-detection dataset. Then, based on YOLOv7, the network was optimized as follows: (1) a CBAM attention module was added to the neck section; (2) a Gaussian receptive field-based label-assignment strategy was implemented at the junction between the original feature-fusion module and the detection head; (3) the SPPFCSPC module was used to replace the multi-space pyramid pooling. By seamlessly uniting CBAM, RFLAGauss, and SPPFCSPC, we establish a novel collaborative optimization framework. Finally, experimental comparisons revealed that the improved model’s accuracy increased from 92.4% to 94.4%; recall improved from 90.5% to 93.9%; and inference speed increased from 87.2 frames per second to 94.2 frames per second. Compared with single-stage object-detection models such as YOLOv7 and YOLOv8, the model demonstrated superior accuracy and inference speed. Its inference speed also significantly outperforms that of Faster R-CNN, Mask R-CNN, DINOv2, and RT-DETRv2, markedly enhancing both small-object (head) detection performance and efficiency. Full article

In underwater signal processing, accurate time delay estimation (TDE) is of crucial importance for ensuring the reliability of data transmission. However, the complex propagation of sound waves and strong noise interference in the underwater environment make this task extremely challenging. Especially under the condition of low signal-to-noise ratio (SNR), the existing methods based on cross-correlation and deep learning struggle to meet requirements. Aiming at this core issue, this paper proposed an innovative solution. Firstly, a multi-sub-window reconstruction is performed on the frequency-sliding generalized colorboxpinkcross-correlation (FS-GCC) matrix between signals to capture the time delay characteristics from different frequency bands and conduct the enhancement and extraction of features. Then, the grayscale image corresponding to the generated FS-GCC matrix is used, and the multi-level noise features are extracted by the multi-layer convolution of denoising convolutional neural network (DnCNN), effectively suppressing the noise and improving the estimation accuracy. Finally, the model-agnostic meta-learning (MAML) framework is introduced. Through training tasks under various SNR conditions, the model is enabled to possess the ability to quickly adapt to new environments, and it can achieve the desired estimation accuracy even when the number of underwater training samples is limited. Simulation validation was conducted under the NOF and NCS underwater acoustic channels, and results demonstrate that our proposed approach exhibits lower estimation errors and greater stability compared with existing methods under the same conditions. This method enhances the practicality and robustness of the model in complex underwater environments, providing strong support for the efficient and stable operation of underwater sensor networks. Full article

As China’s urbanisation continues, the building area is expanding, of which the occupancy of rural residential buildings is also very large. However, most rural dwellings lack insulating structures and have poor thermal performance. This paper verifies and analyses the energy-saving potential of green façades for rural houses in northern China through comparative experiments as well as software simulations. The experiments were conducted from July to August 2024 to verify the reliability of the software simulations. And the simulation was carried out on an existing house in rural northern China. The experimental results show that the reference room consumes 1.84 times more electricity than the vertical greenery room, and the vertical greenery achieves a good energy saving of 45.75%. According to the simulated data, the building energy efficiency of rural houses in northern China after green façade retrofitting is obvious, the energy-saving rate reaches 14.94%, and 713.32 KWh of electricity can be saved in the whole cooling period. Full article