Search Results (3,212)

The increasing penetration of renewable energy sources (RESs) into medium-voltage (MV) and low-voltage (LV) power systems presents significant challenges in ensuring power grid stability and energy sustainability. Advanced power conversion technologies are essential to mitigate voltage and frequency fluctuations while meeting stringent power quality standards. RES-based generation systems typically employ multistage power electronics to achieve: (i) maximum power point tracking; (ii) galvanic isolation and voltage transformation; (iii) high-quality power injection into the power grid. In this context, this paper provides a comprehensive review of up-to-date isolated DC–DC converter topologies tailored for the integration of RES. As a contribution to support this topic, recent advancements in solid-state transformers (SSTs) are explored, with particular emphasis on the adoption of wide bandgap (WBG) semiconductors technologies, such as silicon carbide (SiC) and gallium nitride (GaN). These devices have revolutionized modern power systems by enabling operation at a higher switching frequency, enhanced efficiency, and increased power density. By consolidating state-of-the-art advancements and identifying technical challenges, this review offers insights into the suitability of power converter topologies in light of future trends, serving as a valuable resource for optimizing grid-connected RES-based sustainable power systems. Full article

Low Earth Orbit (LEO) constellations support the transmission of various communication services and have been widely applied in fields such as global Internet access, the Internet of Things, remote sensing monitoring, and emergency communication. With the surge in traffic volume, the quality of user services has faced unprecedented challenges. Achieving accurate low Earth orbit constellation network traffic prediction can optimize resource allocation, enhance the performance of LEO constellation networks, reduce unnecessary costs in operation management, and enable the system to adapt to the development of future services. Ground networks often adopt methods such as machine learning (support vector machine, SVM) or deep learning (convolutional neural network, CNN; generative adversarial network, GAN) to predict future short- and long-term traffic information, aiming to optimize network performance and ensure service quality. However, these methods lack an understanding of the high-dynamics of LEO satellites and are not applicable to LEO constellations. Therefore, designing an intelligent traffic prediction model that can accurately predict multi-service scenarios in LEO constellations remains an unsolved challenge. In this paper, in light of the characteristics of high-dynamics and the high-frequency data streams of LEO constellation traffic, the authors propose a DST-LEO satellite-traffic prediction model (a dynamic spatio-temporal low Earth orbit satellite traffic prediction model). This model captures the implicit features among satellite nodes through multiple attention mechanism modules and processes the traffic volume and traffic connection/disconnection data of inter-satellite links via a multi-source data separation and fusion strategy, respectively. After splicing and fusing at a specific scale, the model performs prediction through the attention mechanism. The model proposed by the authors achieved a short-term prediction RMSE of 0.0028 and an MAE of 0.0018 on the Abilene dataset. For long-term prediction on the Abilene dataset, the RMSE was 0.0054 and the MAE was 0.0039. The RMSE of the short-term prediction on the dataset simulated by the internal low Earth orbit constellation business simulation system was 0.0034, and the MAE was 0.0026. For the long-term prediction, the RMSE reached 0.0029 and the MAE reached 0.0022. Compared with other time series prediction models, it decreased by 22.3% in terms of the mean squared error and 18.0% in terms of the mean absolute error. The authors validated the functions of each module within the model through ablation experiments and further analyzed the effectiveness of this model in the task of LEO constellation network traffic prediction. Full article

Ovarian cancer stands out as one of the most formidable adversaries in women’s health, largely due to its typically subtle and nonspecific early symptoms, which pose significant challenges to early detection and diagnosis. Although existing diagnostic methods, such as biomarker testing and imaging, can help with early diagnosis to some extent, these methods still have limitations in sensitivity and accuracy, often leading to misdiagnosis or missed diagnosis. Ovarian cancer’s high heterogeneity and complexity increase diagnostic challenges, especially in disease progression prediction and patient classification. Machine learning (ML) has outperformed traditional methods in cancer detection by processing large datasets to identify patterns missed by conventional techniques. However, existing AI models still struggle with accuracy in handling imbalanced and high-dimensional data, and their “black-box” nature limits clinical interpretability. To address these issues, this study proposes SHAP-GAN, an innovative diagnostic model for ovarian cancer that integrates Shapley Additive exPlanations (SHAP) with Generative Adversarial Networks (GANs). The SHAP module quantifies each biomarker’s contribution to the diagnosis, while the GAN component optimizes medical data generation. This approach tackles three key challenges in medical diagnosis: data scarcity, model interpretability, and diagnostic accuracy. Results show that SHAP-GAN outperforms traditional methods in sensitivity, accuracy, and interpretability, particularly with high-dimensional and imbalanced ovarian cancer datasets. The top three influential features identified are PRR11, CIAO1, and SMPD3, which exhibit wide SHAP value distributions, highlighting their significant impact on model predictions. The SHAP-GAN network has demonstrated an impressive accuracy rate of 99.34% on the ovarian cancer dataset, significantly outperforming baseline algorithms, including Support Vector Machines (SVM), Logistic Regression (LR), and XGBoost. Specifically, SVM achieved an accuracy of 72.78%, LR achieved 86.09%, and XGBoost achieved 96.69%. These results highlight the superior performance of SHAP-GAN in handling high-dimensional and imbalanced datasets. Furthermore, SHAP-GAN significantly alleviates the challenges associated with intricate genetic data analysis, empowering medical professionals to tailor personalized treatment strategies for individual patients. Full article

The accurate detection and monitoring of levee cracks is critical for maintaining the structural integrity and safety of flood protection infrastructure. Yet at present the application of using UAV to achieve an automatic, rapid detection of levee cracks is still limited and there is a lack of effective deblurring methods specifically tailored for UAV-based levee crack images. In this study, we present E2S, a novel two-stage framework specifically designed for UAV-based levee crack segmentation, which leverages an unsupervised deblurring technique to enhance image quality. In the first stage, we introduce an Improved CycleGAN model that mainly performs motion deblurring on UAV-captured images, effectively enhancing crack visibility and preserving crucial structural details. The enhanced images are then fed into the second stage, where an Attention U-Net is employed for precise crack segmentation. The experimental results demonstrate that the E2S framework significantly outperforms traditional supervised models, achieving an F1-score of 81.3% and a crack IoU of 71.84%, surpassing the best-performing baseline, Unet++. The findings confirm that the integration of unsupervised image enhancement can substantially benefit downstream segmentation tasks, providing a robust and scalable solution for automated levee crack monitoring. Full article

Automatic page layout generation is a challenging and promising research task, which improves the design efficiency and quality of various documents, web pages, etc. However, the current generation of layouts that are both reasonable and aesthetically pleasing still faces many difficulties, such as the shortcomings of existing methods in terms of structural rationality, element alignment, text and image relationship processing, and insufficient consideration of element details and mutual influence within the page. To address these issues, this article proposes a Transformer-based Generative Adversarial Network (TGAN). Generative Adversarial Networks (GANs) innovatively introduce the self-attention mechanism into the network, enabling the model to focus more on key local information that affects page layout. By introducing conditional variables in the generator and discriminator, more accurate sample generation and discrimination can be achieved. The experimental results show that the TGAN outperforms other methods in both subjective and objective ratings when generating page layouts. The generated layouts perform better in element alignment, avoiding overlap, and exhibit higher layout quality and stability, providing a more effective solution for automatic page layout generation. Full article

Face recognition (FR) is a less intrusive biometrics technology with various applications, such as security, surveillance, and access control systems. FR remains challenging, especially when there is only a single image per person as a gallery dataset and when dealing with variations like pose, illumination, and occlusion. Deep learning techniques have shown promising results in recent years using VAE and GAN, with approaches such as patch-VAE, VAE-GAN for 3D Indoor Scene Synthesis, and hybrid VAE-GAN models. However, in Single Sample Per Person Face Recognition (SSPP FR), the challenge of learning robust and discriminative features that preserve the subject’s identity persists. To address these issues, we propose a novel framework called AD-VAE, specifically for SSPP FR, using a combination of variational autoencoder (VAE) and Generative Adversarial Network (GAN) techniques. The proposed AD-VAE framework is designed to learn how to build representative identity-preserving prototypes from both controlled and wild datasets, effectively handling variations like pose, illumination, and occlusion. The method uses four networks: an encoder and decoder similar to VAE, a generator that receives the encoder output plus noise to generate an identity-preserving prototype, and a discriminator that operates as a multi-task network. AD-VAE outperforms all tested state-of-the-art face recognition techniques, demonstrating its robustness. The proposed framework achieves superior results on four controlled benchmark datasets—AR, E-YaleB, CAS-PEAL, and FERET—with recognition rates of 84.9%, 94.6%, 94.5%, and 96.0%, respectively, and achieves remarkable performance on the uncontrolled LFW dataset, with a recognition rate of 99.6%. The AD-VAE framework shows promising potential for future research and real-world applications. Full article

Herein, we present a novel method for junction temperature monitoring of GaN HEMT devices to achieve real-time temperature perception at different locations on the device surface. Through sputtering patterned Ti/Pt thermistor strips on the surface of a GaN HEMT device to construct an on-chip array junction temperature monitoring unit, the thermal distribution of the device during operation is fully reflected. The developed temperature monitoring unit exhibited a desirable temperature coefficient of resistance of 0.183%/°C in the range of 25 °C to 205 °C. Comparison with the thermal imager shows that the integrated temperature monitoring unit can accurately reflect the real-time temperature with a monitoring accuracy of more than 95%, which helps to improve the long-term reliability of GaN power devices under actual application conditions of high frequency and high power density. Full article

Recognizing human activities from motion data is a complex task in computer vision, involving the recognition of human behaviors from sequences of 3D motion data. These activities encompass successive body part movements, interactions with objects, or group dynamics. Camera-based recognition methods are cost-effective and perform well under controlled conditions but face challenges in real-world scenarios due to factors such as viewpoint changes, illumination variations, and occlusion. The latter is the most significant challenge in real-world recognition; partial occlusion impacts recognition accuracy to varying degrees depending on the activity and the occluded body parts while complete occlusion can render activity recognition impossible. In this paper, we propose a novel approach for human activity recognition in the presence of partial occlusion, which may be applied in cases wherein up to two body parts are occluded. The proposed approach works under the assumptions that (a) human motion is modeled using a set of 3D skeletal joints, and (b) the same body parts remain occluded throughout the whole activity. Contrary to previous research, in this work, we address this problem using a Generative Adversarial Network (GAN). Specifically, we train a Convolutional Recurrent Neural Network (CRNN), whose goal is to serve as the generator of the GAN. Its aim is to complete the missing parts of the skeleton due to occlusion. Specifically, the input to this CRNN consists of raw 3D skeleton joint positions, upon the removal of joints corresponding to occluded parts. The output of the CRNN is a reconstructed skeleton. For the discriminator of the GAN, we use a simple long short-term memory (LSTM) network. We evaluate the proposed approach using publicly available datasets in a series of occlusion scenarios. We demonstrate that in all scenarios, the occlusion of certain body parts causes a significant decline in performance, although in some cases, the reconstruction process leads to almost perfect recognition. Nonetheless, in almost every circumstance, the herein proposed approach exhibits superior performance compared to previous works, which varies between $2.2\%$ and $37.5\%$, depending on the dataset used and the occlusion case. Full article

X-ray computed tomography (CT) is vital for medical diagnostics, but frequent radiation exposure raises concerns, driving the adoption of low-dose CT (LDCT) to mitigate risks. However, LDCT often introduces noise, compromising diagnostic accuracy. This paper proposes a pure vision transformer (PViT) for LDCT denoising, enhanced with a gradient–Laplacian attention module (GLAM) to improve edge preservation and fine structural detail reconstruction. The model’s robustness was validated across five diverse datasets (piglet, head, abdomen, chest, thoracic), demonstrating consistent performance in preserving anatomical structures. Extensive ablation studies on attention configurations and loss functions further substantiated the contributions of each module. Quantitative evaluation using PSNR and SSIM, alongside radiologist assessment, confirmed significant noise suppression and sharper anatomical boundaries, particularly in regions with fine details such as organ interfaces and bone structures. Additionally, in benchmark comparisons against state-of-the-art LDCT models (RED-CNN, TED-Net, DSC-GAN, DRL-EMP) and traditional methods (BM3D), the model exhibited lower parameter and stable training performance. These findings highlight the model’s robustness, efficiency, and clinical applicability, making it a promising solution for improving LDCT image quality while maintaining computational efficiency. Full article

To address the challenges posed by complex backgrounds and the low occurrence in photovoltaic cell images captured by industrial sensors, we propose a novel defect detection method: MRA-YOLOv8. First, a multi-branch coordinate attention network (MBCANet) is introduced into the backbone. The coordinate attention network (CANet) is incorporated to mitigate the noise impact of background information on the detection task, and multiple branches are employed to enhance the model’s feature extraction capability. Second, we integrate a multi-path feature extraction module, ResBlock, into the neck. This module provides finer-grained multi-scale features, improving feature extraction from complex backgrounds and enhancing the model’s robustness. Finally, we implement alpha-minimum point distance-based IoU (AMPDIoU) to the head. This loss function enhances the accuracy and robustness of small object detection by integrating minimum point distance-based IoU (MPDIoU) and Alpha-IoU methods. The results demonstrate that MRA-YOLOv8 outperforms other mainstream methods in detection performance. On the photovoltaic electroluminescence anomaly detection (PVEL-AD) dataset, the proposed method achieves a mAP₅₀ of 91.7%, representing an improvement of 3.1% over YOLOv8 and 16.1% over detection transformer (DETR). On the SPDI dataset, our method achieves a mAP₅₀ of 69.3%, showing a 2.1% improvement over YOLOv8 and a 6.6% improvement over DETR. The proposed MRA-YOLOv8 also exhibits great deployment potential. It can be effectively integrated with drone-based inspection systems, allowing for efficient and accurate PV plant inspections. Moreover, to tackle the issue of data imbalance, we propose generating synthetic defect data via generative adversarial networks (GANs), which can supplement the limited defect samples and improve the model’s generalization ability. Full article

Developing reliable noninvasive diagnostic and monitoring systems for diabetes remains a significant challenge, especially in the e-healthcare domain, due to computational inefficiencies and limited predictive accuracy in current approaches. The current study integrates machine learning with a molecularly imprinted polymer biosensor for detecting D-glucose in the exhaled breath condensate or aerosol. Advanced models, such as Convolutional Neural Networks and Recurrent Neural Networks, were used to analyze resistance signals, while classical algorithms served as benchmarks. To address challenges like data imbalance, limited samples, and inter-sensor variability, synthetic data generation methods like Synthetic Minority Oversampling Technique and Generative Adversarial Networks were employed. This framework aims to classify clinically relevant glucose levels accurately, enabling non-invasive diabetes monitoring. Full article

Thermal accumulation under high output power densities is one of the most significant challenges for GaN power devices. Diamond, with its ultra-high thermal conductivity, offers great potential for improving heat dissipation in high-power GaN devices. In this study, nanocrystalline diamond (NCD) passivated high-electron mobility transistors (HEMTs) based on AlGaN/GaN-on-Si heterostructures were fabricated with a gate length of 2 μm. The NCD film has a thickness of 250–383 nm and a uniform morphology with a grain size of mostly ~240 nm. Compared to the devices without NCD passivation, those devices with the NCD passivation layer show an increase in current density from 447 mA/mm to 555 mA/mm, a reduction in on-resistance from 20 Ω·mm to 13 Ω·mm, and a noticeable suppression of current degradation at high-drain voltages. Junction temperature measurements under varied output power densities reveal a 36% improvement in heat dissipation efficiency with the NCD passivation. These results fully demonstrate the promising potential of NCD for enhancing heat dissipation in high-power GaN devices. Full article

This study developed a novel domain-adaptive neural network framework, CNDAD—Net, for addressing the challenges of lung lesion detection in cross-domain medical image analysis. The proposed framework integrates domain adaptation techniques into a classical encoding–decoding structure to align feature distributions between source and target domains. Specifically, a “Generative Adversarial Network” GAN-based domain discriminator is utilized for the iterative refinement of feature representations to minimize cross-domain discrepancies and improve the generalization capability of the model. In addition, a novel Cross-Fusion Block (CFB) is proposed to implement multi-scale feature fusion that facilitates the deep integration of 2D, 3D, and domain-adapted features. The CFB achieves bidirectional feature flow across dimensions, thereby improving the model’s capability to detect diverse lesion morphologies while minimizing false positives and missed detections. For better detection, coarse-grained domain adaptation is implemented by MMD for further optimization. It integrates a module inspired by a CycleGAN for the process to generate high-resolution images on low-quality data. Using the Lung Nodule Analysis (LUNA16) dataset, the test was conducted and its experimental result was compared with that of previous standard methods such as Faster R-CNN and YOLO, yielding mAP 0.889, recall at 0.845 and the F1-score at 0.886. This work, with a novel CNDAD—Net model, lays down a solid and scalable framework for the precise detection of lung lesions, which is extremely critical for early diagnosis and treatment. The model has prospects and is capable of being extended in future to multimodal imaging data ad real-time diagnostic scenarios, and can help in further developing intelligent medical image analysis systems. Full article

With the wide application of the gallium nitride (GaN) preparation method based on Metal–Organic Chemical Vapor Deposition (MOCVD), the automation of MOCVD equipment has become a research hotspot. This paper explores the automation scheme of MOCVD reaction chamber cleaning to improve productivity and reduce labor costs. Firstly, this paper establishes the kinematic solution model of a MOCVD cleaning robot and designs the cleaning robot path planning control algorithm. Considering the error between the initial position of the robot end-effector and the desired initial position in practical applications, this paper further designs a fault-tolerant motion planning algorithm for the initial position error. The simulation results show that the method can effectively reduce the initial position error and make it converge exponentially to zero. Finally, this paper builds the robot control system of the cleaning system and verifies the cleaning effect through tests. The test results show that the system can meet the actual use requirements and realize the reaction chamber cleaning automation goal. Full article

GaN-based terahertz (THz) Schottky barrier diodes (SBDs) are critical components for achieving high-power performance in THz frequency multipliers. However, the space applications of GaN-based THz SBDs are significantly constrained due to insufficient research on the effects of space irradiation. This work investigates the effects of 450 MeV Kr swift heavy ion (SHI) irradiation on the electrical characteristics and induced defects in GaN-based THz SBDs. It was found that the high-frequency performance of GaN-based THz SBDs is highly sensitive to Kr SHI irradiation, which can be attributed to defects induced in the GaN epitaxial layer by the irradiation. Low-frequency noise analysis reveals trap states located at an energy level of approximately 0.62 eV below the conduction band. Moreover, the results from SRIM calculation and photoluminescence spectra confirmed the presence of irradiation-induced defects caused by Kr SHI irradiation. Full article