To mitigate secondary damage caused by reclosing a distribution line with a permanent fault, a highly sensitive fault property identification scheme using detection-based reclosing is proposed in this paper. Firstly, the analytical model for single-phase detection-based reclosing in distribution networks is established to reveal the differences between transient and permanent faults. Given that the equivalent fault resistance of permanent ground faults differs significantly from the insulation resistance of transient faults, an equivalent fault resistance calculation method is proposed to improve the sensitivity of fault property identification. Furthermore, to reduce the impact of fault resistance and load capacity on phase-to-phase fault property identification sensitivity, a permanent fault identification method utilizing the phase-to-phase impedance ratio is proposed. Finally, a comprehensive detection-based reclosing scheme is developed. Validation results based on PSCAD simulations and RTDS closed-loop tests demonstrate that the proposed scheme exhibits both wide applicability across various scenarios and high sensitivity.

We present a practical O(1)-per-query method for non-uniform scalar quantisation when decision boundaries are static and reused across large query volumes. Real-valued boundaries are demoted to an FP12 finite universe (U=212) and indexed using a compact van Emde Boas–style predecessor structure, yielding predecessor retrieval via a constant number of table lookups independent of the number of bins. We show that the coarse lookup is corrected by a bounded local check against neighbouring original boundaries, recovering the exact bin. Experiments show up to a 5× speedup versus binary search for high-volume lookup workloads.

Deep learning-based synthetic aperture radar (SAR) ship image processing is valuable for civilian and military applications, yet training deep models requires abundant high-resolution SAR ship images. Due to the high cost of SAR sensors and limited data availability, constructing large-scale, high-quality datasets remains challenging. SAR image generation offers a promising solution. However, most existing generation methods focus on natural images, with few targeting SAR ship targets. This paper proposes a high-resolution SAR ship image generation method using an improved Denoising Diffusion Generative Adversarial Network (DDGAN). To enhance detail generation, a Context-Aware Upsampling Filter (CAUF) is designed to incorporate SAR-specific information, enabling the backbone network to capture more effective features during upsampling. Experiments show that our method achieves a Fréchet Inception Distance (FID) of 55.21 and a Structural Similarity Index (SSIM) of 0.7245, producing high-quality SAR-consistent images with rich details.

Range sidelobe modulation (RSM) limits the performance of pulse-agile radars. This letter proposes an expectation–maximization (EM)-based minimum mean-square error (MMSE) estimator that adaptively learns the clutter and target distributions and adjusts a range–Doppler (RD) prior map. By optimizing the shrinkage strength via prior weighting, the proposed method improves RSM suppression in the region of interest. Simulation results validate the effectiveness of the proposed approach.

Abstract This study evaluates the operational performance, environmental impact, and regulatory compliance of the WUPA Sludge Treatment Plant, with a focus on sludge and water quality management. Sludge and water samples from the WUPA River were analysed for key physicochemical parameters, including pH, turbidity, dissolved oxygen, and heavy metal concentrations. The results indicate high pollutant removal efficiency, with turbidity, suspended solids, and biochemical oxygen demand reduced by 66.27%, 91.42%, and 91.43%, respectively. While the treated sludge demonstrates suitability for agricultural application, nitrate concentrations exceed permissible limits for direct water discharge. Total dissolved solids removal and electrical conductivity reduction exhibited minor inefficiencies, resulting in an overall plant performance efficiency of 57.73%. Environmental impact assessment revealed minimal risk of leaching or odour to surrounding areas; however, turbidity and trace metal concentrations in river water exceeded WHO drinking water guidelines. The study highlights the need for enhanced nitrate reduction, heavy metal monitoring, and process optimisation to improve both operational efficiency and environmental compliance. Additionally, the findings provide insights into potential energy recovery from sludge and offer transferable strategies for UK wastewater utilities to advance low-carbon, net-zero, and sustainable sludge management practices.

The sensing-based semi-persistent scheduling (SB-SPS) mechanism in 5G New Radio (NR) vehicle-to-everything (V2X) Mode 2 suffers from wireless blind spots (WBS) caused by hidden terminals and half-duplex constraints, which can lead to persistent collisions in dense traffic. This letter proposes a resource map sharing service (RMSS), where a roadside unit (RSU) aggregates a wider-area resource-occupancy map and shares it with vehicles to filter SB-SPS candidates. We analyze interface-dependent delivery of the resource map sharing message (RMSM): PC5 enables low-latency bitmap sharing, while Uu supports a bitmap with reselection counter (RC) information to compensate for map staleness at vehicle user equipment (VUE) under higher latency.

Aiming at the problems of low monitoring accuracy of tool condition and insufficient intelligence of health management, takes the CNC machining center milling titanium alloyas the research object and proposes a tool life cycle health management method based on Digital Twin. By constructing a four-layer digital twin architecture, the bidirectional real-time interaction between the physical machine tool and the virtual model is realized. Through the fusion of multi-source sensor data, a hybrid prediction model based on MSCNN-BiLSTM network is developed, combined with cross-scale dual attention mechanism, which effectively improves the accuracy of tool wear state recognition. The experimental results show that the proposed method can achieve a wear state recognition accuracy of 99.72% in titanium alloy milling, with a root mean square error and mean absolute error of the remaining life of 0.9537 and 0.7860, respectively. Compared with the traditional method, the prediction accuracy is improved. At the same time, the digital twin system developed based on Unity3D realizes the visualization monitoring and intelligent early warning function of tool condition. The results show that the system improves the overall equipment efficiency by 15% to 20%, providing an effective solution for tool health management in the intelligent manufacturing environment.

The increasing penetration of inverter-based resources fundamentally changes the nature of frequency stability in low-inertia power systems. Frequency support becomes predominantly transient-dominated and is governed by the dynamic interaction of converter control structures rather than by aggregate inertia alone. While grid-forming (GFM) inverters are widely regarded as essential for fast frequency support, the dynamic limitations and shaping potential of grid-following (GFL) control architectures remain insufficiently understood. This paper presents a systematic and experimentally backed comparison of inverter-based frequency support concepts based on their transient frequency–power dynamics. GFL droop control, GFM droop control, and Virtual Synchronous Machine (VSM) control are implemented on identical inverter hardware using a common inner control structure. The dynamic mapping from frequency deviation to active power injection is characterized through laboratory experiments and low-order system identification. In addition, an H∞-based matching controller is introduced to shape the closed-loop frequency–power dynamics of a GFL inverter to emulate GFM behavior. The results demonstrate that classical GFL droop control exhibits delayed power responses due to measurement-based synchronization and cascaded tracking loops, leading to deeper frequency nadirs in weak grids. GFM strategies enable immediate power injection and explicit damping, with inertia and damping acting as complementary tuning parameters. Crucially, the experiments show that GFL inverters equipped with H∞ matching control can reproduce key GFM frequency–power characteristics, including rapid initial power injection and substantially improved frequency nadir, in both strong-grid and weak-grid scenarios. These findings indicate that fast frequency support is not inherently tied to the GFM or GFL classification of an inverter, but to the closed-loop frequency–power dynamics achieved by its control structure. Accordingly, a strict dichotomy between GFM and GFL concepts is not sufficient to assess frequency-support capability.

This letter proposes a precision-improved sampling switch that requires only two additional transistors and one inverter, with all devices implemented using N-type transistors. The design introduces a source-drain shorted dummy transistor to compensate for or cancel out the charge injection from the main switch, and incorporates an additional discharge switch with a negative power supply bias to suppress charge leakage and switch soft conduction phenomena. Based on a 7 μm channel length IZO TFT process, simulation results demonstrate that the proposed sampling switch achieves 0-10V full-swing input and output under VDD=10V and a sampling rate of 20kHz, the proposed switch achieves a 0-10 V full-swing input and output. At an input signal frequency of 107.42 Hz, the signal-to-noise and distortion ratio (SNDR) reaches 77.98 dB. The spurious-free dynamic range (SFDR) is improved by 33.36 dB. This design offers an effective approach to improve the sampling accuracy of TFT integrated circuits.

To enhance the fault handling capacity of low-voltage DC (LVDC) distribution networks and mitigate the detrimental effects of faults on converter apparatus, this research introduces an innovative modular fault suppression and handling strategy. The proposed approach entails the integration of multi-port fault limiting modules (FLM) with DC circuit breakers (DCCBs) and DC transformers, thereby enhancing the system’s current and voltage regulation. This integration allows for the use of economical mechanical DCCBs while improving the fault-ride-through (FRT) capabilities of the conversion equipment. Under normal operating conditions, the FLM are bypassed to reduce system disturbances; in the event of a fault, it is engaged to curtail fault currents and preserve voltage stability. The validity of the proposed method is verified through simulations in PSCAD/EMTDC. Simulation results demonstrate that the proposed strategy effectively reduces the peak fault current and limits the transient voltage compared to conventional schemes. Furthermore, the system successfully achieves a fault ride-through by maintaining the fault current within a safe range of the rated current, ensuring robust protection and rapid recovery.

Digitalisation enables the clean energy transition but introduces fragmentation across platforms, protocols, and standards. The PARMENIDES project addresses these challenges by introducing EMS4HESS – an ontology-based Energy Management System for Hybrid Energy Storage Systems within Renewable Energy Communities. EMS4HESS leverages the PARMENI-DES Energy Community Ontology (PECO) to ensure semantic interoperability across data models and information layers. PECO describes community structures, asset properties, optimisation strategies, and flexibility signals, enabling automated reasoning and adaptive control. The system integrates diverse storage technologies and optimises their operation to improve resilience and sustainability. Real-world pilots in Austria and Sweden validate the approach under contrasting regulatory frameworks. Initial results show significant improvements in self-sufficiency and self-consumption compared to baseline scenarios without HESS, confirming the feasibility of ontology-driven, grid-aware energy management as a scalable solution for the clean energy transition.

This study introduces a methodology for categorizing Finnish electricity distribution system operators (DSOs) using geographic spatial data to be used for improved network business regulation. By employing open geographic spatial data and k-means clustering, the study provides a rapid, transparent and unbiased method for categorization. The clustering is implemented to find similar DSOs. The k-means clustering method grouped DSOs into seven categories, revealing more nuanced classifications beyond the traditional three classes. The results were validated by comparing differences in results of another clustering based on the key-figures. The differences were explained by further analysing the results with information of population density. The results are coherent with earlier studies and suggest that geographic spatial data can effectively enable further studies and regulatory applications by bridging datasets from different publishers based on spatial data. Essential for further studies are internal variations within operating environments of DSOs for targeted network business regulation.

In modern communication systems, signal bandwidth continues to expand and the signal environment grows increasingly dense. Accurate extraction of narrowband target signals from wide frequency bands has thus become a key requirement in digital signal processing. Traditional serial digital down conversion (DDC) systems face real-time constraints in high-speed sampling scenarios. They struggle to adapt to the synchronous processing of multiple signals within wide frequency bands. Based on the Xilinx VU5P Field-Programmable Gate Array (FPGA), this paper designs and implements a 16-channel parallel DDC architecture. It adopts parallel mixing and polyphase filtering decimation technology to decompose 5 Gsps high-speed sampling signals into 16 channels of 312.5 Msps low-speed sub-signals. The system can synchronously process 5 MHz narrowband QPSK and 50 MHz wideband QPSK signals under dual carriers of 1 GHz and 1.1 GHz. Simulation and hardware test results indicate that its processing speed is 16 times higher than that of traditional serial architectures. The output signals are free from aliasing distortion, with stopband attenuation of no less than 60 dB and resource occupancy rate below 15%, showing promising engineering application prospects.

Graph Neural Networks (GNNs) have shown promise in modelling complex graph-structured data within power distribution systems. However, their application remains limited by the absence of uncertainty quantification (UQ) techniques, which are crucial for risk-aware decision-making. This paper presents a deep ensemble framework for GNNs that generates uncertainty estimates for n-1 contingency criterion classification predictions, an important task for distribution system operators (DSOs) to evaluate network reliability. Through evaluation of ensemble sizes up to 20 members using bootstrap sampling methodology, we demonstrate that ensemble configurations achieve substantial performance improvements over single models. Across the full 0-100\% range of confidence-based filtering thresholds, the ensemble maintains consistently higher accuracy than a single GNN. The framework provides confidence levels for predictions, enabling data-driven risk management for DSOs. This capability establishes ensemble GNNs as a reliable tool for uncertainty-aware decision-making in power system operations with validated performance guarantees.

This paper addresses issues such as power transmission interruption during rectifier-side faults and susceptibility to system overvoltage during inverter-side faults in wind power DC transmission systems. A hybrid topology comprising a rectifier-side MMC and an inverter-side LCC-MMC configuration is proposed, along with corresponding enhanced control strategies. The characteristics of the proposed hybrid HVDC under bilateral AC faults are investigated. Results indicate that during single-phase faults on the rectifier side, the proposed model and control strategy can maintain transmitted power at 0.54p.u. or higher, preventing power transmission interruptions. Compared to conventional control strategies for inverter-side hybrid DC transmission systems, the minimum power transmission value increases by 11%. Under inverter-side AC faults, wind power output can track power variations during the disturbance. During a single-phase-to-ground fault, the DC voltage of the receiving-end MMC is limited to within 1.13p.u., and during a three-phase short-circuit fault, it is restricted to within 1.29p.u., remaining below the safety threshold of 1.3p.u. Consequently, the proposed system effectively ensures the long-distance and reliable transmission of wind power, making it suitable for the dependable delivery of large-scale offshore wind power.

With the global energy transition and the development of smart manufacturing, modern power systems are characterized by a high penetration of renewable energy and power electronic devices, leading to a declining resilience against voltage sags. Accurate post-event source identification and event inversion estimation form the foundation for managing and mitigating voltage sag issues, which is of great significance for enhancing high-quality power supply in the grid. This paper proposes a method for voltage sag source identification and inversion estimation based on disturbance energy and pattern matching, aiming to achieve precise localization and full-process inversion estimation of voltage sag events, thereby providing key technical support for refined grid management and high-quality power supply. First, by analyzing the changes in energy flow direction at each monitoring point before and after a fault, upstream and downstream identification of the sag source is achieved. Combined with judgment results from multiple monitoring points, the exact location of the voltage sag is accurately determined, overcoming the challenge of insufficient source identification accuracy caused by multi-directional power flow from renewable energy sources in active looped systems. Second, pattern matching technology is introduced to construct a voltage sag event inversion estimation model. Using the Monte Carlo simulation method, a voltage sag source identification pattern library is generated. The measured data from limited monitoring points are intelligently matched with the pre-generated pattern library based on characteristics, enabling post-event inversion of voltage sag events and determining the voltage sag severity at unmonitored nodes. Finally, the IEEE 30-bus system is used as a case study to simulate and verify the effectiveness and accuracy of the proposed method in precise voltage sag source localization and inversion estimation.

This work investigates improving dielectric resonator antenna (DRA) radiation bandwidth through the integration of linearly graded permittivity profiles. Using Characteristic Mode Analysis (CMA), it is shown that such profiles broaden modal distribution, consequently extending the bandwidth of the dominant resonant mode. The proposed antenna design is enabled by additive manufacturing techniques, which allow precise control over material composition and geometry. Experimental results show a significant improvement in both modal and impedance bandwidth from 9.1% to 24,1%, leading to a more stable gain across the operating frequency range.

Plant leaf classification is a core task in biodiversity monitoring, ecological studies, and smart agriculture. In prac-tice, it often still depends on field identification by botanists, which can be slow, laborious, and limited by the availabil-ity of skilled personnel. To address these issues, we present Leaf-StarNet, a compact convolutional neural network tai-lored for accurate and efficient leaf recognition. The network combines three complementary components: an LS-Conv module for localized and efficient feature extraction, an scSE attention block to refine both spatial and channel infor-mation, and a Frequency-Based Enhancement (FBE) unit to better capture subtle frequency-domain patterns. We evalu-ate the model on the LeafSnap dataset and achieve an accuracy of 96.83%, surpassing lightweight baselines such as MobileNetV2, GhostNetV3, and FasterNet-T1. Ablation experiments confirm the contribution of each module, while Grad-CAM visualizations illustrate how the network’s focus shifts from fine textures in early layers to higher-level se-mantic regions in deeper layers. With its small footprint and low computational demand, Leaf-StarNet is well suited for deployment on mobile or embedded platforms, providing a practical and reliable solution for automated plant leaf clas-sification.