Search Results (1,012)

As the operation modes and maintenance management requirements of power grids become increasingly complex, the optimization of monthly outage maintenance schedules faces challenges such as difficulty in multi-department coordination, high reliance on manual experience, and insufficient adaptability to dynamic demands. To address these issues, this paper proposes a collaborative optimization method for monthly outage schedules of power grids based on a large language model and reinforcement learning. The method uses the large language model to parse natural language requirements raised in balance meetings, converting requirements such as fixed maintenance dates, duration adjustments, and forbidden maintenance periods into structured constraint parameters. Subsequently, the small reinforcement learning model based on Dueling Deep Q-Network (Dueling DQN) re-optimizes the outage schedule according to the updated environment parameters, achieving a collaborative solution from meeting requirement parsing to schedule generation. Case study results on the IEEE 39-bus system show that the proposed method can integrate the dynamic requirements arising from multiple rounds of balance meetings and perform adaptive optimization of the outage schedule under different constraint conditions. Experimental results on the IEEE 39-bus system demonstrate that the proposed framework effectively optimizes monthly outage schedules. Compared with the initial schedule, the proposed method reduces monthly voltage violations by 31 times and decreases active power loss by 35.88 MWh. Full article

This paper studies a cooperative wireless system in which a single-antenna base station (BS) communicates with a destination user (U) via a half-duplex energy-harvesting amplify-and-forward relay, while the direct BS–U link is unavailable. The destination (U) is equipped with a fluid antenna system (FAS) comprising multiple closely spaced receive ports, enabling spatial reconfigurability through instantaneous port selection. A power-splitting architecture is adopted at the relay to support simultaneous energy harvesting and information forwarding. All wireless links are modeled as flat Rayleigh fading, and the spatial correlation among the FAS ports is explicitly incorporated. To analytically characterize the impact of correlated port selection, a Gaussian copula framework is employed to model the joint distribution of the FAS-channel power gains. Exact integral expressions for the cumulative distribution function of the end-to-end signal-to-noise ratio are derived, from which the outage probability is obtained. For the special case of uncorrelated FAS ports, closed-form expressions are further developed using order statistics and special functions. In addition, asymptotic analysis is carried out to provide further insight into system performance in the high-signal-to-noise-ratio region. Numerical and Monte Carlo simulation results validate the analytical derivations and demonstrate that FAS-based receiver selection yields significant gains in outage performance, even in the presence of strong spatial correlation and energy-harvesting constraints. Full article

Long-term and large-scale power outages (blackouts) pose a significant threat to the functioning of modern cities, exposing the level of their preparedness for crisis situations. A blackout is a systemic phenomenon whose cascading impacts extend far beyond the energy sector, affecting multiple interdependent domains of urban operations. Therefore, this study aims to evaluate the resilience of urban systems to electricity supply disruptions across infrastructural, institutional, social, and functional dimensions, based on blackout case studies from selected cities in Poland, Spain, and Portugal. Utilising data spanning the period 2008–2025, a qualitative cross-case comparison was conducted based on indicator coding and the identification of recurrent patterns. The results indicate that a blackout simultaneously impacts all four dimensions of urban resilience. The highest vulnerability was observed within the social and functional dimensions, manifested by the limited preparedness of residents for prolonged power deprivation, as well as disruptions in information flow and municipal service delivery. Furthermore, the findings reveal that a high level of infrastructural resilience does not guarantee effective crisis response without adequate institutional and social support. Based on these insights, it is recommended to develop a European Union-level framework to secure financial resources aimed not only at mitigating the consequences of blackouts—which frequently exhibit cross-border characteristics—but primarily at funding proactive prevention and preparedness measures against power grid failures. Full article

Due to the continuously rising demand for electric vehicle use in distribution networks and power system disruptions caused by natural events and cyber threats, difficulties arise in managing, designing, and recovering the active distribution network. This paper proposes a whole process restoration model considering an electric vehicle battery swapping station with black-start service and islanding partition with loop-elimination radiality. Firstly, the framework for electric vehicle demand and the charging and discharging processes of battery swapping facilities is created, reflecting driving patterns and the order of charging and discharging. Next, the possible main bus and the removal of circular networks are considered in the context of multi-timeframe islanding partitions. Then, the models of units’ start-up sequence and grid reconstruction are established to ensure the system’s successful restoration. The objective function is to minimize the outage loss, restoration time, switch operations and scale of distribution islands. Finally, the practicality and efficiency of the suggested approach are confirmed through the PG&E69-bus system and the 185-node distribution network, which achieves considerable enhancement in system strength and dependability. Full article

Reliable power systems are essential to modern life, as severe storms continue to disrupt grid stability and cause widespread outages. Predicting storm outages enables utilities and emergency managers to pre-stage resources and improve resilience. However, the several days of forecast lead time typically needed for preparedness significantly affect the accuracy of outage predictions. This study investigates the impact of forecast lead time on the error propagation of a Gradient Boosting Machine (GBM)-based outage prediction model (OPM) driven by Weather Research and Forecasting (WRF) model forecasts and analysis predictions. We evaluate three error-analysis scenarios: FFAP (forecast vs. analysis-based outage predictions), FFAO (forecast vs. actual outages), and LFAO (leave-one-storm-out forecast vs. actual outages). Model performance is compared using Mean Absolute Percentage Error (MAPE) and Centered Root-Mean-Square Error (CRMSE) across short (12 h–1 d), medium (2–3 d), and long (4–5 d) forecast lead-time categories, with the long category representing the upper end of the medium-range forecast window relevant to operational preparedness. The results show that forecast lead time substantially affects outage prediction accuracy, but the magnitude depends on the evaluation setup. In the controlled FFAP scenario, CRMSE increased by approximately 110% as lead time increased, from 259 to 543 outages, isolating the effect of weather forecast degradation. In the more operational LFAO scenario, CRMSE was already high at short lead times, increasing from 847 to 920 outages, indicating that model generalization error dominates once storms are unseen. Across scenarios, LFAO errors were 51% higher than FFAO errors at short lead times, highlighting the importance of testing outage models under unseen-event conditions. These results quantify how forecast degradation and model generalization jointly shape the reliability of outage prediction and provide practical guidance for lead-time-aware storm preparedness. Full article

Industrial Internet of Things (IIoT) networks use on-demand sensing and wireless power transfer (WPT) for self-sustainable operation. Existing scheduling frameworks are fundamentally limited because they react only after energy levels decline. Consequently, IoT nodes enter charging mode only when their residual energy falls below a threshold, causing energy outages, increased latency, and missed sensing tasks while preventing proactive WPT resource allocation. This paper proposes a Digital Twin (DT)-enabled proactive scheduling framework that transforms IIoT scheduling from reactive to proactive. The key innovation is a closed-loop virtual–real integration in which a DT layer, co-located with the control centre, maintains a Kalman filter predictor to forecast node energy over an H-slot horizon, enabling scheduling decisions before energy shortages occur. Physical layer security (PLS) constraints and DT-based anomaly detection protect against eavesdropping, energy depletion, and false data injection attacks. A multi-objective formulation jointly optimises sensing utility and WPT efficiency while accounting for DT synchronisation overhead and uplink bandwidth consumption. The resulting multi-slot Binary Integer Linear Programmes (BILP) are solved using branch-and-bound with a reliability branching rule, and a fast greedy heuristic is also developed. Simulation results over 50 Monte Carlo iterations show that the proposed framework reduces energy outage events by approximately 70% compared with the reactive baseline, activates less than 50% of available sensing nodes, and schedules less than 60% of energy transmitters for WPT. Ablation studies confirm that DT prediction is the primary contributor to the outage reduction. DT-based anomaly detection achieves a false alarm rate below 3% while maintaining a detection rate above 95%. The proposed framework improves the sustainability, efficiency, and security of IIoT networks with practical computational overhead, making it well suited for Industry 5.0 deployments. Full article

Electricity load forecasting is essential for balancing energy supply and demand, reducing energy waste, and maintaining power grid stability. Accurate forecasts enable power utilities to optimize energy dispatch and mitigate the risk of supply shortages or outages. However, conventional forecasting methods often struggle to capture highly nonlinear local fluctuations in electricity consumption and long-term temporal dependencies. To address these challenges, this study proposes MSCNN-ResLSTM, a hybrid model for multi-step electricity load forecasting. The proposed model integrates Multi-Scale Convolutional Neural Networks (MSCNNs) to extract local time-series features at multiple temporal scales, residual blocks (ResBlocks) to enhance feature representation through residual connections, and Long Short-Term Memory (LSTM) networks to model long-range temporal dependencies. To comprehensively evaluate its effectiveness, a cross-paradigm experimental framework is established in which MSCNN-ResLSTM is compared with seven representative benchmark models from three methodological categories: traditional machine learning (Extreme Gradient Boosting-XGBoost), classical recurrent and convolutional neural networks (LSTM, Temporal Convolutional Network-TCN, CNN-LSTM, MSCNN-LSTM, and Direct LSTM (Seq2Seq)), and self-attention-based architectures (Transformer). Experimental results show that MSCNN-ResLSTM achieves higher forecasting accuracy and greater stability across the full 24-step prediction horizon, consistently outperforming all competing baselines while effectively suppressing recursive error propagation. Full article

Reliable electric-power restoration is important for autonomous ships because propulsion, navigation, communication, and emergency loads depend on a compact shipboard distribution network with limited generation redundancy. This paper studies the fault reconfiguration of an autonomous shipboard meshed microgrid under generator outage, branch fault, and dynamic-load disturbance conditions. A multi-objective model is established by considering priority-based load restoration, switching-operation cost, and generator load balancing. To represent emergency load management more realistically, a continuous restoration ratio is introduced for aggregated shipboard load groups, so that full restoration, derated operation, and load shedding can be described in one formulation. A Q-learning-guided ant colony optimization method (QL-ACO) is then proposed. In this method, Q-learning is used as an adaptive parameter controller for the pheromone factor, heuristic factor, and greedy selection probability, rather than as a direct switch-action selector. Elite reinforcement and pheromone smoothing are also introduced to reduce premature convergence. Four shipboard fault scenarios are simulated, including a single-branch fault, a single-generator outage, a combined branch–generator fault, and a dynamic-load–branch-fault case. The results show that the proposed method maintains critical-load restoration, improves Class-III load recovery in complex scenarios and obtains feasible reconfiguration schemes with fewer switching operations than fixed-parameter ACO, NSGA-II, PSO, and a compact direct RL reference baseline. Runtime, scalability, statistical, and sensitivity analyses are also provided to examine online applicability and robustness. Full article

Accurate prediction of the Remaining Useful Life (RUL) of nuclear power plant systems can support more informed maintenance strategies and help reduce unplanned outages, motivating continued research into reliable, physically grounded prognostic models. This study evaluates machine-learning-based prognostic models, focusing on their ability to learn degradation patterns directly from operational data. Baseline Feedforward Neural Networks (FNNs), Long Short-Term Memory (LSTM) networks, Graph Neural Networks (GNNs), and a hybrid GNN–Transformer architecture are assessed using data generated from the ASHERAH dynamic Pressurized Water Reactor simulator, which incorporates realistic degradation mechanisms, including condenser fouling and pump head loss. The proposed hybrid model integrates graph-based representations of component interactions with Transformer-based temporal attention to capture both system-level dependencies and long-term degradation dynamics. Model performance is evaluated using error-based and tolerance-based metrics, including Prediction Within Bounds Accuracy ($PWB{A}_{20\%}$), alongside uncertainty calibration via Prediction Interval Coverage Probability (PICP), with uncertainty quantified through deep ensembles. Results show that baseline NNs exhibit limited predictive accuracy and poor uncertainty calibration, while models incorporating temporal modeling and system topology achieve substantial improvements. The GNN–Transformer attains the strongest performance, yielding the highest $PWB{A}_{20\% }$ (87.3%) and significantly improved uncertainty calibration, with an average PICP of 90.0%. These findings demonstrate the effectiveness of topology-aware, attention-based architectures for robust and reliable RUL prediction in complex nuclear systems. Full article

The medium- and low-voltage distribution network is a critical hub connecting the transmission grid and end-users, and its power supply reliability directly determines livelihood security and socio-economic operational efficiency. Typhoon-induced strong winds and rainfall often trigger large-scale power outage risks, severely threatening power grid security and resilience. To achieve the rapid and accurate perception of outage risk areas based on building damage after typhoons, this study proposes the ResiDS-Net method, which infers distribution network outage risk levels by identifying building damage levels. An improved Dempster–Shafer evidence theory is here adopted to fuse the outputs of CM-ResNet50, Inception-V3 and DenseNet121, enhancing perception accuracy. A two-stage “coarse screening–fine judgment” framework using dual datasets is established to quickly identify large-scale suspected power outage areas from building group damage data. To address the issue that equating building damage with power outages reduces judgment accuracy, this study further develops a hierarchical building damage dataset, classifying individual buildings by damage level to achieve precise outage risk identification. Our experiments show that ResiDS-Net achieves 96.66% and 93.00% accuracy on the two datasets, 2.13% and 2.50% higher than nine comparative networks including Inception-V3. The proposed method effectively improves outage risk perception precision and provides a scientific basis for power emergency repair. It should be noted that the proposed method provides building-damage-based outage risk inference for emergency decision support, rather than the direct detection of verified actual outage status. Full article

High penetration of renewable energy sources has intensified voltage fluctuations and violations in modern distribution networks. Although distributed control offers a scalable solution, existing methodologies frequently require the exchange of sensitive operational data. This critical limitation, which is intrinsic to distributed algorithms, raises significant privacy concerns. This paper proposes a novel privacy-preserving distributed voltage regulation scheme based on the framework of state-based potential games. Specifically, we first formulate a potential function that aligns the local objectives of individual buses with the voltage profile improvement goal. To reduce the risk of sensitive-data disclosure, we introduce a decoupling mechanism where buses only exchange information regarding coupling constraint violations rather than bus voltage or power injection data. Furthermore, a parallel update law is established, allowing buses to optimize their strategies independently. A key strength of the proposed scheme is its resilience to stochastic communication outages (SCOs). Numerical tests demonstrate that, in comparison to existing distributed voltage control methods, the proposed scheme not only reduces the risk of sensitive-data disclosure but also maintains highly consistent performance across diverse SCO scenarios. Finally, the effectiveness and convergence performance of the proposed voltage regulation scheme are validated by the case studies. Full article

Electric vehicle (EV) usage has increased significantly in the past few years, which may create challenges for distribution system operators due to EV charging needs. In this paper, we propose an approach for planning and operating EV charging stations, considering the challenges that distribution networks may face. A two-step framework is proposed in this paper. First, the optimal size and location of a charging station is determined using a multi-objective optimization problem considering minimizing power losses and voltage drop while maximizing load placements. Then, an optimal scheduling scheme is employed to charge and discharge the vehicles on the selected buses. Simulation studies were conducted using IEEE 33- and 123-bus systems; the results show that the proposed framework significantly enhances the buses’ voltages and line power flows. In order to plan for charging stations, several factors need to be considered, such as optimal size and location, the daily load curve for the given system, the time of use (TOU), and the charging patterns of EV owners. Without careful planning and operation, the system may suffer vulnerability and line overloading, which may lead, eventually, to cascading outages and interruptions. By improving grid utilization, reducing losses, and enabling coordinated EV charging and discharging, the proposed framework supports more sustainable energy use and facilitates the integration of electric mobility into future low-carbon power systems. Full article

This paper introduces a methodology for evaluating the voltage stability margin in high-voltage direct-current (HVDC) systems, which analyzes the singularity of the power flow Jacobian matrix—computed via the Newton—Raphson method—and identifies saddle-node bifurcations. The continuation power flow method is employed to model progressive load increases, enabling the continuous tracking of power flow solutions and the determination of voltage collapse points. Within this framework, the system’s behavior is analyzed under contingency conditions, particularly transmission line outages, assessing its capability to maintain secure operating conditions under increasing demand scenarios. The main objective is to identify the most critical line in the system, defined as that which leads to the greatest reduction in loadability when unavailable, prior to voltage collapse. This approach allows for the early identification of structural vulnerabilities, supporting decision-making processes aimed at risk mitigation and operating cost optimization. The proposed methodology is validated using two systems: the six-terminal CIGRE-B4 HVDC system and an 11-node HVDC test feeder. Full article

Short-term fluctuations in dissolved oxygen are difficult to capture in warm outdoor Nile tilapia (Oreochromis niloticus) ponds using periodic manual measurements, yet they strongly influence fish performance and farm management. This study presents a preliminary field evaluation of a low-cost IoT workflow for dissolved oxygen monitoring and short-horizon forecasting in pond-based tilapia culture. An ESP32-based sensing node continuously measured dissolved oxygen, temperature, and pH, transmitted readings to a cloud backend, and generated short-horizon forecasts from 5 min aggregated windows. During live validation from 1 to 10 April 2026, the 30 min forecast achieved a mean absolute error of 0.783 mg/L and directional accuracy of 60.23%, with only modest improvement over a persistence baseline. The 6 h forecast achieved 1.109 mg/L and 53.82%, respectively, indicating limited predictive value at the extended horizon. An extended 47-day field deployment (May–June 2026) captured four sensor-recorded low-DO events and two documented power outages, causing sensor downtime and providing additional field-deployment evidence. These results demonstrate the engineering feasibility of the integrated workflow, but they do not establish robust operational forecasting validity because the data were collected from one pond, high-frequency records were temporally correlated, and independent reference-meter validation was not available. The study is, therefore, best interpreted as a proof-of-concept field evaluation that identifies practical requirements for future low-cost aquaculture forecasting systems. Full article

This paper presents innovative research on power grid reliability, which is essential for the overall sustainability of energy systems in the emerging age of electricity. The research primarily analyzes the profound methodological disparities between the fragmented European approach and the exact statistical model employed in the United States (the 2.5 Beta method defined by the IEEE 1366 standard). A novel dimension addressed in this research is the compounding effect of climate hazards and the massive proliferation of artificial intelligence (AI) data centers. Unlike conventional single-layer machine learning models (such as standard Support Vector Machines or regressions) that rely solely on historical weather data, this study proposes the Hierarchical Spatiotemporal Multiplex Networks (HMN-RTS) predictive framework. By dynamically fusing structured environmental data with unstructured social sensor data (Geographic Information Systems—GIS, and social media feeds), the proposed HMN-RTS framework significantly outperforms traditional models in predicting outage risks and their exact durations. Full article