Search Results (71)

We address the challenge that occlusions in on-branch soybean images impede accurate pod-level phenotyping. We propose a lab on-branch pipeline that couples a prior-guided synthetic data generator (producing synchronized visible and amodal labels) with an amodal instance segmentation framework based on an improved Swin Transformer backbone with a Simple Attention Module (SimAM) and dual heads, trained via three-stage transfer (synthetic excised → synthetic on-branch → few-shot real). Guided by complete (amodal) masks, a morphology-driven module performs pose normalization, axial geometric modeling, multi-scale fused density mapping, marker-controlled watershed, and topological consistency refinement to extract seed per pod (SPP) and geometric traits. On real on-branch data, the model attains Visible Average Precision (AP) 50/75 of 91.6/77.6 and amodal AP50/75 of 90.1/74.7, and incorporating synthetic data yields consistent gains across models, indicating effective occlusion reasoning. On excised pod tests, SPP achieves a mean absolute error (MAE) of 0.07 and a root mean square error (RMSE) of 0.26; pod length/width achieves an MAE of 2.87/3.18 px with high agreement (R² up to 0.94). Overall, the co-designed data–model–task pipeline recovers complete pod geometry under heavy occlusion and enables non-destructive, high-precision, and low-annotation-cost extraction of key traits, providing a practical basis for standardized laboratory phenotyping and downstream breeding applications. Full article

Triboelectric nanogenerators (TENGs) have emerged as efficient mechanical-energy harvesters with advantages—simple architectures, broad material compatibility, low cost, and strong environmental tolerance—positioning them as key enablers of self-powered systems. This review synthesizes recent progress in energy-storage interfaces, power management, and system-level integration for TENGs. We analyze how intrinsic source characteristics—high output voltage, low current, large internal impedance, and pulsed waveforms—complicate efficient charge extraction and utilization. Accordingly, this work highlights a variety of power-conditioning approaches, including advanced rectification, multistage buffering, impedance transformation/matching, and voltage regulation. Moreover, recent developments in the integration of TENGs with storage elements, cover hybrid topologies and flexible architectures. Application case studies in wearable electronics, environmental monitoring, smart-home security, and human–machine interfaces illustrate the dual roles of TENGs as power sources and self-driven sensors. Finally, we outline research priorities: miniaturized and integrated power-management circuits, AI-assisted adaptive control, multimodal hybrid storage platforms, load-adaptive power delivery, and flexible, biocompatible encapsulation. Overall, this review provides a consolidated view of state-of-the-art TENG-based self-powered systems and practical guidance toward real-world deployment. Full article

Biological hydrogels are widely available in terms of raw material sources and can be processed and molded using relatively simple techniques. Hydrogels can offer abundant three-dimensional, water-containing channels that facilitate the reaction between gases and dye, making them the preferred choice for the solid support layer in colorimetric indicators. However, biomass hydrogels exhibit inferior mechanical properties, making them unsuitable for large-scale manufacturing processes. In this study, four dual-network composite hydrogels Agar/Gelatin, Sodium Alginate/Agar, Sodium Alginate/Poly (vinyl alcohol), Sodium Alginate/Gelatin (AG/Gel, SA/AG, SA/PVA and SA/Gel) prepared through hybrid strategies. Furthermore, the influence of the dual-network structure on the mechanical properties and ammonia response was systematically investigated, using microscopy and Fourier transform infrared spectroscopy (FTIR) characterization method. The experimental results demonstrate that the incorporation of SA into original hydrogel matrices can significantly enhance both the mechanical and ammonia response performance due to the secondary topological network structure. The interpenetrating double network structure was effectively regulated through the calcium ion cross-linking process. The color difference threshold of SA/PVA’s response to ammonia gas is 10, it holds promise for rapid detection applications. The SA/Gel composite hydrogel exhibits excellent mechanical robustness and toughness. The tensile strength of the SA/Gel sample is 11 times that of a single gel, and the toughness is 80 times greater, suggesting its suitability for large-scale manufacturing of colorimetric indicator. Full article

Under the background of “dual-carbon” strategy and global energy transition, the metallurgical industry, which accounts for 15–20% of industrial energy consumption, urgently needs to reduce the energy consumption and emission of DC power supply of electric furnaces. Aiming at the existing 400–800 V/≥3000 A industrial-frequency transformer-rectifier system with low efficiency, large volume, heat dissipation difficulties and other bottlenecks, this thesis proposes and realizes a high-frequency integrated DC power supply scheme for high-power electric furnaces: high-frequency transformer core and rectifier circuit are deeply integrated, which breaks through and reduces the volume of the system by more than 40%, and significantly reduces the iron consumption; multiple cores and three windings in parallel are used for the system. The topology of multiple cores and three windings in parallel enables several independent secondary stages to share the large current of 3000 A level uniformly, eliminating the local overheating and current imbalance; the combination of high-frequency rectification and phase-shift control strategy enhances the input power factor to more than 0.95 and cuts down the grid-side harmonics remarkably. The authors have completed the design of 100 kW prototype, magneto-electric joint simulation, thermal structure coupling analysis, control algorithm development and field comparison test, and the results show that the program compared with the traditional industrial-frequency system efficiency increased by 12–15%, the system temperature rise reduced by 20 K, electrode voltage increased by 10–15%, the input power of furnace increased by 12%, and the harmonic index meets the requirements of the traditional industrial-frequency system. The results show that the efficiency of this scheme is 12–15% higher than the traditional IF system, the temperature rise in the system is 20 K lower, the voltage at the electrode end is 10–15% higher, the input power of the furnace is increased by 12%, and the harmonic indexes meet the requirements of GB/T 14549, which verifies the value of the scheme for realizing high efficiency, miniaturization, and reliable DC power supply in metallurgy. Full article

Autonomous navigation in dynamic, multi-vessel maritime environments presents a formidable challenge, demanding strict adherence to the International Regulations for Preventing Collisions at Sea (COLREGs). Conventional approaches often struggle with the dual imperatives of global path optimality and local reactive safety, and they frequently rely on simplistic state representations that fail to capture complex spatio-temporal interactions among vessels. We introduce a novel hybrid reinforcement learning framework, D* Lite + Transformer-Enhanced Soft Actor-Critic (TE-SAC), to overcome these limitations. This hierarchical framework synergizes the strengths of global and local planning. An enhanced D* Lite algorithm generates efficient, long-horizon reference paths at the global level. At the local level, the TE-SAC agent performs COLREGs-compliant tactical maneuvering. The core innovation resides in TE-SAC’s synergistic state encoder, which uniquely combines a Graph Neural Network (GNN) to model the instantaneous spatial topology of vessel encounters with a Transformer encoder to capture long-range temporal dependencies and infer vessel intent. Comprehensive simulations demonstrate the framework’s superior performance, validating the strengths of both planning layers. At the local level, our TE-SAC agent exhibits remarkable tactical intelligence, achieving an exceptional 98.7% COLREGs compliance rate and reducing energy consumption by 15–20% through smoother, more decisive maneuvers. This high-quality local control, guided by the efficient global paths from the enhanced D* Lite algorithm, culminates in a 10–32 percentage point improvement in overall task success rates compared to state-of-the-art baselines. This work presents a robust, verifiable, and efficient framework. By demonstrating superior performance and compliance with rules in high-fidelity simulations, it lays a crucial foundation for advancing the practical application of intelligent autonomous navigation systems. Full article

The photovoltaic, energy storage, direct current, and flexibility (PEDF) system represents a crucial innovation for transforming buildings into low-carbon energy sources. Although it is still in the early stages of scalable demonstration, current research and practice related to PEDF lack comprehensive studies on optimizing and evaluating system capacity configuration across various scenarios. Capacity configuration and energy scheduling are crucial components that are often treated separately, leading to a missing opportunity to leverage the synergy among key interactive devices. To address this issue, this paper proposes an optimization and evaluation framework for the PEDF system that employs a dual-layer model for planning and operating. This framework precisely configures the PEDF topology, load, photovoltaic, energy storage, and critical interactive devices, while integrating economic, environmental, and reliability objectives. The effectiveness of the proposed model has been validated in optimizing capacity configurations for newly built office buildings and existing commercial settings. The results indicate that for new office buildings, schemes that prioritize low-carbon initiatives are more effective than those that focus on reliability and economy. In existing commercial buildings, reliability-focused schemes outperform those that prioritize economy and low carbon, and all three are significantly better than pre-configuration schemes. The proposed framework enhances the theoretical understanding of PEDF system planning and evaluation, thereby promoting broader adoption of sustainable energy technologies. Full article

This paper introduces a novel dual-stream collaborative architecture for remote sensing road segmentation, designed to overcome multi-scale feature conflicts, limited dynamic adaptability, and compromised topological integrity. Our network employs a parallel “local–global” encoding scheme: the local stream uses depth-wise separable convolutions to capture fine-grained details, while the global stream integrates a Swin-Transformer with a graph-attention module (Swin-GAT) to model long-range contextual and topological relationships. By decoupling detailed feature extraction from global context modeling, the proposed framework more faithfully represents complex road structures. Comprehensive experiments on multiple aerial datasets demonstrate that our approach outperforms conventional baselines—especially under shadow occlusion and for thin-road delineation—while achieving real-time inference at 31 FPS. Ablation studies further confirm the critical roles of the Swin Transformer and GAT components in preserving topological continuity. Overall, this dual-stream dynamic-fusion network sets a new benchmark for remote sensing road extraction and holds promise for real-world, real-time applications. Full article

Modular multilevel converter (MMC)-based power electronic transformers (PETs) present a promising solution for connecting AC/DC microgrids to facilitate renewable energy access. However, the capacitor ripple voltage in MMC-based PET submodules hinders volume optimization and power density enhancement, significantly limiting their application in distribution networks. To address this issue, this study introduces an optimized method for suppressing the submodule capacitor ripple voltage in MMC-based PET systems under normal and grid fault conditions. First, an MMC–PET topology featuring upper and lower arm coupling is proposed. Subsequently, a double-frequency circulating current injection strategy is incorporated on the MMC side to eliminate the double-frequency ripple voltage of the submodule capacitor. Furthermore, a phase-shifting control strategy is applied in the isolation stage of the dual-active bridge (DAB) to transfer the submodule capacitor selective ripple voltages to the isolation stage coupling link, effectively eliminating the fundamental frequency ripple voltage. The optimized approach successfully suppresses capacitor ripples without increasing current stress on the isolated-stage DAB switches, even under grid fault conditions, which are not addressed by existing ripple suppression methods, thereby reducing device size and cost while ensuring reliable operation. Specifically, the peak-to-peak submodule capacitor ripple voltage is reduced from 232 V to 10 V, and the peak current of the isolation-stage secondary-side switch is limited to ±90 A. The second harmonic ripple voltage on the LVDC bus can be decreased from ±5 V to ±1 V with the proposed method under the asymmetric grid voltage condition. Subsequently, a system simulation model is developed in MATLAB/Simulink. The simulation results validated the accuracy of the theoretical analysis and demonstrated the effectiveness of the proposed method. Full article

Recent advances in deep learning have witnessed the wide application of convolutional neural networks (CNNs), Transformer models, and Mamba models in optical remote sensing image (ORSI) analysis, particularly for salient object detection (SOD) tasks in disaster warning, urban planning, and military surveillance. Although existing methods improve detection accuracy by optimizing feature extraction and attention mechanisms, they still face limitations when dealing with the inherent challenges of ORSI. These challenges mainly manifest as complex backgrounds, extreme scale variations, and topological irregularities, which severely affect detection performance. However, the deeper underlying issue lies in how to effectively align and integrate local detail features with global semantic information. To tackle these issues, we propose the Trans-Mamba Hybrid Network with Semantic Feature Alignment (TSFANet), a novel architecture that exploits intrinsic correlations between semantic information and detail features. Our network comprises three key components: (1) a Trans-Mamba Semantic-Detail Dual-Stream Collaborative Module (TSDSM) that combines CNNs-Transformer and CNNs-Mamba in a hybrid dual-branch encoder to capture both global context and multi-scale local features; (2) an Adaptive Semantic Correlation Refinement Module (ASCRM) that leverages semantic-detail feature correlations for guided feature optimization; and 3) a Semantic-Guided Adjacent Feature Fusion Module (SGAFF) that aligns and refines multi-scale semantic features. Extensive experiments on three public RSI-SOD datasets demonstrate that our method consistently outperforms 30 state-of-the-art approaches, effectively accomplishing the task of salient object detection in remote sensing imagery. Full article

While deep learning-based image matching methods excel at extracting high-level semantic features from remote sensing data, their performance degrades significantly under cross-daylight conditions and wide-baseline geometric distortions, particularly in multi-source street-view scenarios. This paper presents a novel illumination-invariant framework that synergistically integrates geometric topology and semantic consistency to achieve robust multi-view matching for cross-daylight urban perception. We first design a self-supervised learning paradigm to extract illumination-agnostic features by jointly optimizing local descriptors and global geometric structures across multi-view images. To address extreme perspective variations, a homography-aware transformation module is introduced to stabilize feature representation under large viewpoint changes. Leveraging a graph neural network with hierarchical attention mechanisms, our method dynamically aggregates contextual information from both local keypoints and semantic topology graphs, enabling precise matching in occluded regions and repetitive-textured urban scenes. A dual-branch learning strategy further refines similarity metrics through supervised patch alignment and unsupervised spatial consistency constraints derived from Delaunay triangulation. Finally, a topology-guided multi-plane expansion mechanism propagates initial matches by exploiting the inherent structural regularity of street scenes, effectively suppressing mismatches while expanding coverage. Extensive experiments demonstrate that our framework outperforms state-of-the-art methods, achieving a 6.4% improvement in matching accuracy and a 30.5% reduction in mismatches under cross-daylight conditions. These advancements establish a new benchmark for reliable multi-source image retrieval and localization in dynamic urban environments, with direct applications in autonomous driving systems and large-scale 3D city reconstruction. Full article

Aiming at the limitations of existing multisensor fault diagnosis methods for rolling bearings in real industrial scenarios, this paper proposes an innovative intuitionistic fuzzy weighted least squares twin support higher-order tensor machine (IFW-LSTSHTM) model, which realizes a breakthrough in the noise robustness, adaptability to the working conditions, and the class imbalance processing capability. First, the multimodal feature tensor is constructed: the fourier synchro-squeezed transform is used to convert the multisensor time-domain signals into time–frequency images, and then the tensor is reconstructed to retain the three-dimensional structural information of the sensor coupling relationship and time–frequency features. The nonlinear feature mapping strategy combined with Tucker decomposition effectively maintains the high-order correlation of the feature tensor. Second, the adaptive sample-weighting mechanism is developed: an intuitionistic fuzzy membership score assignment scheme with global–local information fusion is proposed. At the global level, the class contribution is assessed based on the relative position of the samples to the classification boundary; at the local level, the topological structural features of the sample distribution are captured by K-nearest neighbor analysis; this mechanism significantly improves the recognition of noisy samples and the handling of class-imbalanced data. Finally, a dual hyperplane classifier is constructed in tensor space: a structural risk regularization term is introduced to enhance the model generalization ability and a dynamic penalty factor is set to set adaptive weights for different categories. A linear equation system solving strategy is adopted: the nonparallel hyperplane optimization is converted into matrix operations to improve the computational efficiency. The extensive experimental results on the two rolling bearing datasets have verified that the proposed method outperforms existing solutions in diagnostic accuracy and stability. Full article

This article proposes a high-efficiency isolated three-port resonant converter for DC microgrids, combining a dual active bridge (DAB)–LLC topology with hybrid Pulse Width Modulat-Pulse Frequency Modulation (PWM-PFM) phase shift control. Specifically, the integration of a dual active bridge and LLC resonant structure with interleaved buck/boost stages eliminates cascaded conversion losses. Energy flows bidirectionally between ports via zero-voltage switching, achieving a 97.2% efficiency across 150–300 V input ranges, which is a 15% improvement over conventional cascaded designs. Also, an improved PWM-PFM shift control scheme dynamically allocates power between ports without altering switching frequency. By decoupling power regulation and leveraging resonant tank optimization, this strategy reduces control complexity while maintaining a ±2.5% voltage ripple under 20% load transients. Additionally, a switch-controlled capacitor network and frequency tuning enable resonant parameter adjustment, achieving a 1:2 voltage gain range without auxiliary circuits. It reduces cost penalties compared to dual-transformer solutions, making the topology viable for heterogeneous DC microgrids. Based on a detailed theoretical analysis, simulation and experimental results verify the effectiveness of the proposed concept. Full article

High-precision remote sensing image semantic segmentation plays a crucial role in Earth science analysis and urban management, especially in urban remote sensing scenarios with rich details and complex structures. In such cases, the collaborative modeling of global and local contexts is a key challenge for improving segmentation accuracy. Existing methods that rely on single feature extraction architectures, such as convolutional neural networks (i.e., CNNs) and vision transformers, are prone to semantic fragmentation due to their limited feature representation capabilities. To address this issue, we propose a hybrid architecture model called PLGTransformer, which is based on dual-encoder collaborative enhancement and integrates pyramid pooling and graph convolutional network (i.e., GCN) modules. Our model innovatively constructs a parallel encoding architecture combining Swin transformer and CNN: the CNN branch captures fine-grained features such as road and building edges through multi-scale heterogeneous convolutions, while the Swin transformer branch models global dependencies of large-scale land cover using hierarchical window attention. To further strengthen multi-granularity feature fusion, we design a dual-path pyramid pooling module to perform adaptive multi-scale context aggregation for both feature types and dynamically balance local and global contributions using learnable weights. Specifically, we introduce the GCNs to build a topological graph in the feature space, enabling geometric relationship reasoning for multi-scale feature nodes at high resolution. Experiments on the Potsdam and Vaihingen datasets show that our model outperforms contemporary advanced methods and significantly improves segmentation accuracy for small objects such as vehicles and individual buildings, thereby validating the effectiveness of the multi-feature collaborative enhancement mechanism. Full article

Under the rapid advancements in information technology, the complex network characteristics of agricultural product trade relationships among global economies have exhibited increasing prominence. This study takes the soybean trade market as an empirical case, employing a combination of social network analysis to investigate the dynamic evolution of agricultural trade network structures; then, the Temporal Exponential Random Graph Model (TERGM) is adopted to analyse the factors influencing the soybean trade network. Based on comprehensive empirical data encompassing soybean trade data among 126 economies from 2000 to 2022, this research demonstrates several key findings: Firstly, the soybean trade network is characterised by pronounced trade agglomeration effects and “small-world” properties, accompanied by heightened trade substitutability. Secondly, the network’s structural configuration has undergone a distinct transformation, shifting from a traditional single-core–periphery structure to a more complex multi-core–periphery architecture. Thirdly, in response to external shocks impacting network topology, the core structure exhibits greater resilience and stability, whereas the periphery displays heterogeneous responses. Finally, the evolution of soybean trade relations is governed by a dual mechanism involving both endogenous dynamics and exogenous influences. Full article

High-impedance fault detection poses significant challenges for distribution network maintenance and operation. We propose a dual-path neural network for high-impedance fault detection. To enhance feature extraction, we use a Gramian Angular Field algorithm to transform 1D zero-sequence voltage signals into 2D images. Our dual-branch network simultaneously processes both representations: the CNN extracts spatial features from the transformed images, while the GRU captures temporal features from the raw signals. To optimize model performance, we integrate the Crested Porcupine Optimizer (CPO) algorithm for the adaptive optimization of key network hyperparameters. The experimental results demonstrate that our method achieves a 99.70% recognition accuracy on a dataset comprising high-impedance faults, capacitor switching, and load connections. Furthermore, it maintains robust performance under various test conditions, including different noise levels and network topology changes. Full article