Ground-based cloud image categorization is crucial for ground-cloud observation and significantly impacts activities like photovoltaic power forecasting. Current techniques fail to effectively integrate novel cloud image texture features with image classification algorithms. Enhancing cloud categorization precision in publicly available datasets and research tailored for photovoltaic power forecasting remains in its early stages. This study introduces a novel Cloud-ConvNeXt classification approach utilizing a comprehensive texture feature map. We generate a detailed texture feature map for significant areas of ground-based cloud images using a gray-level co-occurrence matrix (GLCM) and local binary patterns (LBP). This image serves as input for an enhanced ConvNeXt classification network, producing precise cloud categories based on meteorological standards. Additionally, cloud occlusion feature and cloud factor feature are extracted from ground-based cloud images to improve cloud categories properties. Results show that the Cloud-ConvNeXt technique surpasses other models, achieving an accuracy of 85.40%. We have constructed a Support Vector Machine (SVM)-Informer model to predict the power output of highly efficient photovoltaic (PV) systems in the next 5 minutes. The results demonstrate that our predictions outperform other methods that depend on cloud images acquired from the ground, with a coefficient of determination (R²) of 0.9668. Furthermore, ablation experiments on attributes linked to different cloud categories confirm their efficacy, thus building a strong foundation for precise solar production forecasting.
28 June 2016 in Protection and Control of Modern Power Systems
In order to overcome the shortages of diagnostic method for distribution networks considering the reliability assessment, this paper proposed a method based on power supply safety standards. It profoundly analyzed the security standard of supply for urban power networks, and established quantitative indicators of load groups based on different fault conditions. Then a method suitable for diagnostic evaluation of urban distribution networks in China was given. In the method, “N-1” calibration analysis of the distribution network was conducted. Then the results are compared with quantitative indicators of load groups on different conditions deriving the diagnostic conclusions and the standard revision is discussed. The feasibility and accuracy of the method is finally verified in the case study.
Rural distribution networks increasingly face voltage quality challenges due to high penetration of distributed renewable energy, heterogeneous rural load behavior, and long radial feeder structures with limited voltage regulation capability. Photovoltaic generation variability and agricultural load fluctuations can lead to voltage rise, reverse power flow, and branch congestion, particularly in weak rural grids. Conventional deterministic voltage control approaches relying on tap changers and capacitor banks often struggle to maintain stable voltage profiles under stochastic operating conditions. This paper proposes a risk-aware coordinated operation framework for rural multi-energy microgrids that integrates stochastic scenario modeling, voltage state perception, and adaptive optimization-based control. Renewable generation uncertainty and rural load variability are represented through correlated scenario generation and Wasserstein-distance-based scenario reduction, where 100 raw joint photovoltaic-load trajectories are reduced to 20 representative scenarios after convergence and distributional-fidelity tests. A stochastic optimization model is developed to coordinate photovoltaic inverters, battery energy storage systems, demand-side flexibility, and reactive compensation devices while satisfying network power-flow, voltage-security, storage, and communication-delay-aware implementation constraints. To mitigate extreme voltage deviation events, the framework incorporates a Conditional Value-at-Risk formulation that penalizes tail-risk voltage violations and maintains voltages within a preferred operating band of – p.u. Case studies on a modified IEEE 33-bus rural distribution system with MW of photovoltaic capacity and MWh of battery storage demonstrate consistent performance improvements across deterministic, risk-neutral stochastic, chance-constrained, and robust baselines. The proposed strategy reduces peak branch loading from in the deterministic benchmark to , while the 95th percentile voltage deviation risk decreases from 0.0071 p.u.2 to 0.0020 p.u.2. Sensitivity, scenario-convergence, scalability, and seasonal representative-day analyses further confirm that the CVaR layer suppresses rare but severe voltage excursions without imposing excessive curtailment or computational burden.
The development of electric vehicles (EVs) reduces dependence on fossil fuels, promotes energy conservation and emissions reduction, and facilitates the transition to clean energy sources in the power grid. However, subjective charging behavior among EV owners can lead to blind charging practices, compromising the reliability of the distribution network by widening the peak-to-valley difference. To address safety concerns during the charging process, this paper proposes hardware and software systems for an experimental verification system. The network architecture, focused on charging safety, is examined. Analysis of the system’s operation data reveals that it enables bidirectional interaction between electric vehicles and the power grid. This solution proves ef-fective for integrating a large number of EVs in peak-shaving and valley-filling efforts, laying a technical foundation for their inclusion in the power grid for peak shaving, valley filling, as well as providing standby and frequency regulation services.
Accurate substation-level Short-Term Load Forecasting (STLF) is essential for secure day-ahead power-system operation, yet localized demand is often affected by meteorological variation and discrete calendar shifts such as statutory holidays and makeup workdays. At this spatial scale, end-to-end forecasting models may over-smooth abrupt local changes and fail to represent peaks and valleys accurately. To address this issue, this study proposes a Calendar-Aware Frequency-Decoupled Framework (CA-FDF) for 24 h ahead substation load forecasting. The load series is decomposed by the Discrete Wavelet Transform (DWT), and the low-frequency component is tracked by a regime-aware Recursive Least Squares (RLS) filter. The residuals are then refined through explicit calendar-state encoding and day-ahead weather forecasts. A Multi-Layer Perceptron (MLP) learns latent weather representations, while SHapley Additive exPlanations (SHAP) interpret calendar- and weather-related effects. Experiments on hourly operational data from 29 anonymized substations in East China show that CA-FDF achieves a Mean Absolute Percentage Error (MAPE) of 1.92% and outperforms representative baselines under the same day-ahead setting. The results indicate that frequency-decoupled residual refinement improves localized load forecasting, with calendar-aware correction contributing the largest gain.