Search Results (521)

Grain-oriented (GO) silicon steel cores in low-voltage current transformers suffer magnetic degradation from residual stress and increased dislocation density during slitting and winding. This study addresses the gap in systematic optimization of secondary annealing on assembled toroidal cores using a 3² full-factorial design varying temperature (650, 850, 1050 °C) and holding time (60, 90, 120 min) on M4 grade cores. Results showed temperature is the dominant factor, while holding time exhibits a synergistic non-linear effect. The optimal condition (850 °C, 90 min) reduced specific losses from 0.85 W/kg to 0.43 W/kg (49% reduction). Mechanistic analysis confirmed this improvement is driven by complete primary recrystallization (equiaxed grains ~50–60 µm), dislocation annihilation (~10 HV hardness reduction), and reinforcement of the Goss texture ({110} <001>). SEM, EDS, and ICP-OES demonstrated that the Carlite coating remained dimensionally (1.67–1.83 µm) and chemically stable, with beneficial decarburization. Temperatures above 850 °C caused magnetic deterioration due to excessive grain growth. These results provide a validated, industrial framework for recovering magnetic efficiency in wound toroidal cores without compromising coating integrity. Full article

Accurate non-invasive voltage measurement is critical for the stable operation of ultra-high-voltage direct-current (UHVDC) grids. In practical converter valve environments, voltage inversion based on the charge simulation method (CSM) may be affected by nearby charged conductors. To address this problem, this paper proposes a non-invasive voltage measurement method combining radially aligned near-conductor two-sensor differential electric-field measurement with three-dimensional electrostatic finite-element modelling. The differential electric field between two radial sensing positions is used for voltage inversion, which suppresses distant common-mode interference. When a nearby interference conductor exists, a weighted differential correction coefficient k is introduced to compensate for the residual radial interference component. Theoretical and simulation results show that k is a scenario-dependent coefficient affected by the measured voltage, sensor spacing, interference voltage, and geometric configuration. In an ultra-high-voltage (UHV) converter valve bridge-arm scenario with a 400 kV interference conductor, the absolute voltage inversion error is reduced from 0.50–1.57% FS before correction to below 0.20% FS after correction. Experiments on a 30 kV-scaled platform further verify the method under different measured voltages, sensor spacings, and interference-voltage levels, with the best-tested case reducing the maximum error from 0.93% FS to 0.16% FS. Full article

As new energy vehicles and energy storage systems become more common, safety accidents caused by lithium-ion batteries overheating have become more of a concern. Early detection based on distinctive gases (such as H₂ and CO) can give an earlier warning than typical monitoring methods like temperature, voltage, or impedance. Nonetheless, attaining high-precision identification in intricate mixed-gas settings continues to be difficult because of the considerable cross-sensitivity of metal oxide semiconductor (MOS) gas sensors. This research presents an ISA-LSTM-TCN multi-task learning model utilizing an enhanced spatial attention mechanism for the swift identification and concentration forecasting of distinctive gases during lithium-ion battery thermal runaway. The model improves key feature extraction and anti-noise performance by combining the long-term temporal modeling ability of the Long Short-Term Memory (LSTM) network with the multi-scale feature extraction ability of the Temporal Convolutional Network (TCN). It also adds an Improved Spatial Attention (ISA) module with a residual multiplication structure. Moreover, in a multi-task learning framework, joint optimization of gas categorization and concentration regression is facilitated using a hard parameter-sharing method. Tests using a built MOS sensor array dataset show that the model is 99.23% accurate at classifying gases and that the $R^2$ values for predicting H₂ and CO concentrations are 0.9510 and 0.8400, respectively. Tests on public datasets and in different noisy environments show that the model is even better at generalizing and is more robust. The results show that the suggested method allows for quick, accurate detection of thermal runaway gases. This makes it an effective and smart way to monitor battery safety warning systems. Full article

This paper investigates abnormal secondary voltage drops in dry-type transformers operating in isolated offshore power systems. While conventional analyses primarily attribute voltage deviations to load conditions and transformer impedance, this study shows that noticeable voltage drops can also occur under no-load conditions due to the combined effects of voltage unbalance, harmonic distortion, and residual magnetic flux. A comprehensive approach integrating on-site measurements, PSCAD simulations, and laboratory experiments is employed to systematically analyze this phenomenon. The results indicate a coupled electromagnetic effect in which source-side voltage imperfections induce asymmetric core flux distribution, which is associated with reduced secondary voltage. In addition, a relationship between synchronous generator winding pitch and harmonic voltage distortion is observed, suggesting its influence on power quality in isolated grids. Simulation results show that the interaction of these factors can lead to a secondary voltage drop of approximately 4–6 V under no-load conditions, even in the absence of transformer defects. Finally, mitigation strategies based on voltage balancing and harmonic reduction are experimentally validated, restoring the secondary voltage to 1.002 pu. These findings provide practical insights for improving voltage stability and power quality in offshore and other isolated power systems. Full article

The dual-active-bridge (DAB) converter is the core component of the DC micro-grid system; it has the advantages of topological structure symmetry, high efficiency, and high-power density. Model predictive control (MPC) is often employed to improve the dynamic response characteristics of the system, but its strong parameter dependence is a key factor limiting the development of MPC. Therefore, a model-free predictive control (MFPC) method combining an ultra-local model with model predictive control is proposed to solve the problem of strong dependence of traditional MPC on system model parameters. Firstly, establish the ultra-local mathematical model of the DAB converter. The system’s lumped disturbances are identified using the residual prediction method and substituted into the discrete model of the system at the next time step to achieve model-free prediction. Secondly, a minimum back-flow power constraint is added to the cost function to improve the steady-state performance of the converter. Thirdly, in the extended phase shift modulation, the Lagrange multiplier method (LMM) is proposed to reduce the current stress, ultimately achieving the collaborative optimization of the comprehensive performance of the DAB. Finally, a simulation model is built using MATLAB/Simulink, and compared with traditional control methods, the voltage ripple has been reduced by 51.3%, 89.1%, and 85.1%, respectively; the current stress significantly decreases both when the output voltage reference value changes and when the load resistance changes abruptly, and both can basically achieve zero back-flow power operation. The validity and superiority of the proposed strategy have been verified. Full article

Accurate demagnetization fault diagnosis is critical to ensuring the safety and reliability of permanent magnet synchronous motors (PMSMs). However, the number, location, and severity of demagnetized permanent magnets are mutually coupled, leading to a combinatorial explosion of fault patterns. Existing methods are largely limited to idealized assumptions involving single-magnet demagnetization or uniform demagnetization of multiple magnets, making it difficult to characterize the random nature of demagnetization in practical operation. Thus, this paper proposes a precise demagnetization fault diagnosis method based on a novel search coil (SC) configuration, in which only two toroidal-yoke-type search coils are installed in the stator slots. The proposed method partitions the rotor permanent magnets into several modules and categorizes the infinite demagnetization fault patterns into 26 representative patterns, effectively addressing the issue of fault mode explosion. Theoretical analysis and experimental results show that the voltage waveforms of the search coil over a single electrical period exhibit significant and stable differences across the identified patterns. By constructing feature vectors based on these differences, a physically interpretable mapping between the feature vectors and fault patterns is established. Combined with a corresponding pattern recognition algorithm, the proposed method enables fast and accurate differentiation of the 26 patterns without the need for complex machine learning models, thereby achieving precise localization of demagnetized permanent magnets. Simulation and experimental results verify the correctness and effectiveness of the proposed method. Full article

Natural disasters and extreme meteorological events are primary causes of unplanned outages in modern power systems. Existing early warning methods suffer from insufficient non-linear feature extraction, severe class imbalance, and limited minority-class recall under single-classifier architectures. This paper proposes a seven-class meteorological fault early warning framework that integrates a sparse autoencoder (SAE), a G1–entropy composite weighting scheme, SMOTE oversampling, and a soft-voting BP–XGBoost ensemble. A leakage-free experimental protocol confines SMOTE exclusively to the training partition, eliminating data contamination from evaluation. Validated on 1955 fault records from a regional grid in East China covering 110 kV, 220 kV, and 500 kV voltage levels (2013–2022), the proposed framework achieved 96.42% accuracy and a 97.46% macro F1-score on the held-out test set, outperforming SVM (72.68%), Random Forest (89.31%), LSTM (81.47%), 1D-CNN (85.38%), and LightGBM (92.15%). Ablation experiments confirmed that SMOTE and G1–entropy weighting contributed macro F1 gains of 8.34 and 6.91 percentage points, respectively, while removing the XGBoost branch degraded accuracy by 28.25%. Temporal validation on 2019–2022 records yielded 91.57% accuracy, confirming temporal generalization. Error analysis further revealed that bidirectional misclassification between lightning damage and wind damage, rooted in shared atmospheric instability signatures, constitutes the dominant residual error source, providing theoretical guidance for future threshold optimization strategies. Full article

To investigate the effect of welding speed on the out-of-plane deformation of Q420 low-alloy high-strength steel thin plates, this study employed a three-dimensional digital image correlation system to monitor the deformation dynamically during TIG welding and cooling. Unlike existing studies that mostly focus on post-weld residual deformation or a single welding stage, this study, under a fixed current of 36 A and arc voltage of 14 V, sets welding speeds ranging from 4.5 to 11.8 mm/s, and for the first time systematically reveals the complete evolution path of Q420 thin plate (2 mm) welding deformation, which includes “thermal expansion—instability mutation—elastic rebound—residual stabilization”. The results show that the welding speed is significantly negatively correlated with the out-of-plane deformation. Although low-speed welding has a high peak plastic strain, the final residual strain is almost completely released; while high-speed welding has a low peak strain but retains a relatively high residual strain. This abnormal phenomenon reveals the deep mechanism that the accumulation and release of plastic strain are asymmetrically regulated by the welding speed. These findings support process optimization for high-strength steel thin plates. Full article

This study investigated the electrical insulation properties of epoxy matrix composites reinforced with carbonized palm kernel shell (PKS) particles. The raw PKS particles were collected, sun-dried, and further oven-dried at 105 °C for 2 h to eliminate residual moisture. The dried shells were then carbonized in an airtight furnace at three different temperatures: 450, 550, and 650 °C. After carbonization, the material was crushed and sieved into particle sizes of 200, 400, and 800 µm using an electromagnetic sieve shaker. Composites were fabricated by incorporating carbonized PKS particles into an epoxy resin matrix at varying weight fractions of 30, 40, 50, and 60 wt%. Electrical insulation performance was evaluated at room temperature and pressure using high-voltage DC test equipment for dielectric strength and a digital insulation tester (MIT 520/2) for resistivity measurements. The results revealed that optimal dielectric strength and resistivity were achieved with smaller particle sizes, lower filler loadings, and at low temperatures. Mineralogical characterization via X-ray diffraction confirmed that there was no radioactive element. Scanning Electron Microscopy revealed porous microstructures within the carbonized particles. Energy-dispersive X-ray spectroscopy indicated that carbon accounted for the highest elemental composition, followed by oxygen. It is concluded that PKS-reinforced epoxy composites exhibit promising electrical insulation properties. Full article

The formation and growth of isolated domains during local switching by a biased tip of a scanning probe microscope in a (100) cut of a bismuth titanate Bi₄Ti₃O₁₂ single crystal were studied experimentally. The as-grown domain structure consists of two domain types: a-type (out-of-plane) and b-type (in-plane). Local switching of the a-type domain area leads to anisotropic growth of a hexagonal a-type domain (a-a switching) with 180° walls. The dependence of the domain size on the pulse duration during domain growth along the b-axis was considered in terms of the anisotropic current-limited domain wall motion. Local switching of the b-type domain area leads to formation of a hexagonal a-type domain (b-a switching) with 90° walls increasing in size linearly with the applied voltage. The dependence of the domain size on the pulse duration was measured over a wide range of humidities. The increase in the domain size at moderate humidity is attributed to the effect of the water meniscus. The decrease in the domain size at high humidity is attributed to backswitching under the action of the residual depolarization field, facilitated by a conductive water layer on the side surfaces of the sample. The obtained results provide useful insights into the domain kinetics of ferroelectrics with C₂ symmetry and can pave the way for the development of domain engineering techniques. The obtained results establish a direct relationship between local switching kinetics, crystallographic anisotropy, and environmental conditions. This provides the scientific community with a new framework for understanding domain wall motion in multiaxial ferroelectrics, which is essential for the development of stable and reliable domain-engineered devices. Full article

Electricity inconsistencies in mini-grids, stemming from meter drift, telemetry faults, topology misconfiguration, non-technical losses, phase imbalance or data manipulation, often emerge as weak, spatially distributed deviations that are difficult to anticipate, yet timely warning is important for future monitoring frameworks in rural electrification and island mini-grids. Existing approaches either apply post hoc threshold-based alarms to individual channels or employ deep learning models that treat metering points independently, ignoring the spatial coupling imposed by the electrical topology and lacking mechanisms to enforce physical feasibility under scarce labeled data. This paper introduces PhysGTT, a Physics-Guided Self-Supervised Graph Temporal Transformer that models the mini-grid as a topology-aware graph and combines a residual Graph Convolutional Network encoder with a temporal Transformer. PhysGTT employs self-supervised pretraining via masked multi-sensor reconstruction and contrastive regime alignment to exploit unlabeled operational data and incorporates gradient-coupled physics regularization through power-balance, voltage-bound and ramp-rate penalties applied to a learned reconstruction head, while producing constraint-level attributions that identify the dominant physical violation pattern for each forecast. PhysGTT is evaluated on a proxy benchmark derived from the UCI Individual Household Electric Power Consumption dataset and on the IEEE 13-node test feeder simulated in OpenDSS and it is compared under identical experimental protocols with eight baselines spanning recurrent, graph-temporal and unsupervised architectures. On the proxy benchmark, PhysGTT achieves an AUC-ROC of 0.8959, an F1-score of 0.8307 and a False Alarm Rate of 0.41%, improving the F1-score by 2.2% relative to the strongest recurrent baseline (GRU) and by up to 15.2% relative to the LSTM baseline, while reducing the False Alarm Rate by approximately 52% relative to the LSTM baseline. On the IEEE 13-node feeder, PhysGTT attains an AUC-ROC of 0.9016 and an F1-score of 0.8361. These results indicate that integrating topology-aware encoding, self-supervised pretraining and physics-guided learning provides a promising and interpretable framework for proactive inconsistency forecasting under synthetic and feeder-simulation benchmarks, although field validation on naturally occurring faults remains necessary. Full article

The proliferation of vehicle-to-grid (V2G) charging stations in distribution networks introduces both voltage regulation challenges and untapped reactive power resources. This paper proposes a reactive–active power coordination control strategy for grid-forming (GFM) V2G charging stations to achieve voltage regulation in radial distribution networks. First, a voltage–reactive power sensitivity matrix is analytically derived from the linearized DistFlow equations, quantifying the voltage influence of each V2G station. The sensitivity matrix is computed from the network topology and line parameters, and its accuracy under varying operating conditions is validated against nonlinear power flow solutions. Second, a dynamic residual reactive capacity model exploits the inverter apparent power margin without curtailing active power, and a sensitivity-weighted proportional allocation distributes the reactive power references among stations. Third, a two-timescale hierarchical control architecture is designed: the upper layer solves a quadratic programming problem every 60 s to determine optimal set-points, while the lower layer employs GFM droop control with a 1 ms response to track references and provide inertia support. Simulation results on a modified IEEE 33-bus system demonstrate that the proposed method reduces the maximum voltage deviation by 62% compared with active-power-only control, while maintaining a frequency nadir of 49.73 Hz, confirming negligible frequency performance degradation. Extended simulations covering a 24 h period with stochastic EV arrival and departure patterns as well as varying load conditions further confirm the robustness of the proposed strategy. Full article

Power flow (PF) analysis is fundamental for power system operation and planning, yet traditional methods like Newton–Raphson face problems in convergence and computational efficiency. While deep learning (DL) offers promising solutions, its “black-box” nature and unstable training dynamics hinder practical adoption. This paper proposes a physics-informed neural network (PINN) framework integrated with a novel half-inverse gradient (HIG) mechanism to address these limitations. First, a systematic study of gradient scaling in PF optimisation found that the lack of enough inverse matrix compensation was the main cause of training instability. Second, we design a residual-driven HIG method that compensates gradient matrices via inverse operations, enabling accelerated convergence while maintaining numerical stability. Third, we develop parameterized voltage variables with differentiable activation functions to enforce hard operational constraints. The HIG optimizer leverages automatic differentiation and truncated singular value decomposition to balance diagonal/non-diagonal gradient information, achieving 99% accuracy in case4gs and case30 studies. Experiments on case118 demonstrate the framework’s scalability, with 65% accuracy compared to about 38% for baseline physics-informed approaches. Full article

A probabilistic data-driven framework for voltage quality forecasting in distribution transformers based on Gaussian process regression and high-resolution field measurements is presented. Voltage time series acquired under real operating conditions were modeled using composite covariance functions designed to capture long-term trends and stochastic multi-scale fluctuations. The proposed approach enables simultaneous prediction and uncertainty quantification, allowing direct compliance assessment with voltage quality standards. The additive Gaussian process models achieved coefficients of determination above 0.75 and produced statistically uncorrelated residuals, indicating an adequate representation of the intrinsic temporal structure. However, the predictive intervals exhibit a certain level of undercoverage, indicating that, while uncertainty is effectively quantified, there is still room for improvement in calibration. The selected kernel structures revealed distinct physical regimes in the voltage dynamics, including smooth steady operation, moderately irregular behavior associated with localized disturbances, and multi-scale stochastic variability. For benchmarking purposes, results were compared with those obtained from a stochastic damped harmonic oscillator with restoring force, a naive model, a seasonal naive model and an Autoregressive Integrated Moving Average model. The oscillator model, the naive model, the seasonal naive model, and the Autoregressive Integrated Moving Average model generated strongly autocorrelated residuals, whereas the Gaussian process models yielded consistent white-noise residuals that outperformed all the other models. These findings demonstrate that probabilistic Gaussian process modeling provides an interpretable, scalable, and uncertainty-aware alternative for predictive voltage quality assessment in modern distribution systems. Full article

The rapid demand for sustainable and efficient energy storage solutions has prompted the pursuit of eco-friendly electrode materials. Biomass-derived carbons from food waste offer a promising pathway to meet this need by combining waste valorization, environmental benefits, and high electrochemical performance. This review highlights that food waste biomass is an effective and inexpensive source of precursors for producing high-performance carbon materials for supercapacitors. Food waste, which includes fruit peels and vegetable residues, cereal husks, and oilseed residues, is a good source of lignocellulosic components, heteroatoms, and structural features that determine the electrochemical characteristics of the derived carbons. These wastes produce hierarchically porous carbons with high surface areas (>1500 m² g⁻¹) on pyrolysis and activation that provide superior ion transport, wettability and pseudocapacitive behaviour. Their electrochemical performance includes capacitances up to 520 F g⁻¹ and energy densities of 35–70 Wh kg⁻¹ in optimized systems, particularly under extended voltage windows or in hybrid supercapacitor configurations, and high cycling stability is equal to or even better than traditional carbons such as activated carbon and graphene. Additionally, biomass valorization contributes to a high level of greenhouse gas capture, decreases landfill, and correlates with the idea of a circular economy. The commercialization potential of biomass-based supercapacitors is supported by recent developments in AI-based optimization, combined with scalable synthesis methods, which would support ecologically, economically, and technologically sustainable energy storage on a large scale. Full article