Short-term Power Research Articles

Abstract Short-term load forecasting is essential for power system stability and resource optimization, but existing methods struggle with load randomness and complexity, resulting in low accuracy. This paper proposes a novel two-phase decomposition-based hybrid forecasting model that integrates Successive Variational Mode Decomposition (SVMD), Spectral Entropy, Ensemble Empirical Mode Decomposition (EEMD), and R-Informer. The proposed method first employs SVMD to decompose the load sequence into multiple intrinsic mode functions (IMFs) of different frequencies. Then, the spectral entropy of each IMF is calculated. The IMF with the highest spectral entropy is further decomposed using EEMD method. This technique yields a set of stationary components. Furthermore, this paper proposes an Adaptive Recursive Cycle (ARC) mechanism, which utilizes learnable periodic embeddings and exponential moving average (EMA) to adaptively model periodic patterns. The ARC mechanism enhances periodic information representation through linear transformation. Based on this, this paper proposes the R-Informer model, which integrates ARC and Informer to predict the decomposed components. Experimental results demonstrate that the proposed forecasting model effectively handles non-stationary load sequences and achieves high accuracy in short-term power load forecasting.

Grid security and system dispatch can be compromised by pronounced volatility in power load under extreme meteorological conditions. However, the dynamic and nonlinear interactions between power load and meteorological variables across diverse weather scenarios are not well captured by existing methods, resulting in limited accuracy and robustness. To address this gap, a short-term power load forecasting model with a dual-channel architecture is proposed. Features are extracted in parallel via dual-channel feature extraction (DCFE): the first channel employs an improved Cascaded Multiscale 2D Convolutional Network (CMCNN) to model local fluctuations and global periodicity in the load time series. The second channel derives scenario-aware variable weights using the Maximal Information Coefficient (MIC); meteorological variables are then gated and weighted before being processed by a multi-layer self-attention network to learn global dependencies. Subsequently, dynamic feature-level fusion is achieved through cross-attention, strengthening key interactions between power load and meteorological factors. The fused representation is fed into an Attention-Enhanced Bidirectional Gated Recurrent Unit (AE-BiGRU) to precisely model temporal dependencies across multiple weather scenarios. Experiments on five years of power load and meteorological data from a region in Australia indicate that the proposed method outperforms the best baseline across multiple weather conditions: RMSE, MAE, MAPE, and sMAPE decrease on average by 32.44%, 31.42%, 30.73%, and 31.05%, respectively, while R2 increases by 0.034 on average, demonstrating strong adaptability and robustness.

Short-term multi-step wind power prediction model based on Pt-Transformer neural network integrating spatio-temporal feature and sparse attention

Transformer-based deep neural networks for short-term solar power prediction in the Middle East and North Africa regions

Short-term wind power forecasting with small-sample datasets using an attention-enhanced domain-adversarial neural network

Short-term power load forecasting for industrial buildings based on decomposition reconstruction and TCN-Informer-BiGRU

Short-term power load forecasting based on parallel decomposition

Short-term power load forecasting for estate-level buildings considering multilevel feature extraction and adaptive fusion

Short-term power prediction of photovoltaic power station based on LSTM-XGBoost model

This article critically analyzes Serbian President Aleksandar Vučić’s political rhetoric, examining his strategic use of nationalist discourse regarding Kosovo, Vojvodina, and alleged “colored revolution” following 2024–2025 student protests. The analysis demonstrates how Vučić manipulates public opinion and consolidates power by positioning himself as Serbia’s sole defender against existential threats. Vučić’s discourse treats Kosovo as Serbia’s “heart,” exploiting deep historical grievances and national identity concerns to reinforce his role as protector against foreign interference. Similarly, he portrays Vojvodina as a new secessionist threat despite no genuine independence movement, expanding his nationalist narrative to rally support. Student-led protests are framed as externally orchestrated “colored revolution” attempts to destabilize Serbia. This perpetual crisis narrative, featuring both internal and external enemies, justifies authoritarian measures while suppressing opposition. Vučić’s rhetoric effectively constructs Serbian national identity while legitimizing his political authority. However, the analysis reveals broader implications for post-conflict states, where such discourse may consolidate short-term power at democracy’s expense. The findings suggest that while nationalist victimhood narratives strengthen Vučić’s immediate political position, they ultimately risk undermining Serbia’s democratic development and institutional progress. 1

Conventional power load forecasting frameworks face limitations in dynamic spatial topology capture and long-term dependency modeling. To address these issues, this study proposes a hybrid GAT-CNN-LSTM architecture for enhanced short-term power load forecasting. The model integrates three core components synergistically: Graph Attention Network (GAT) dynamically captures spatial correlations via adaptive node weighting, resolving static topology constraints; a CNN-LSTM module extracts multi-scale temporal features—convolutional kernels decompose load fluctuations, while bidirectional LSTM layers model long-term trends; and a gated fusion mechanism adaptively weights and fuses spatio-temporal features, suppressing noise and enhancing sensitivity to critical load periods. Experimental validations on multi-city datasets show significant improvements: the model outperforms baseline models by a notable margin in error reduction, exhibits stronger robustness under extreme weather, and maintains superior stability in multi-step forecasting. This study concludes that the hybrid model balances spatial topological analysis and temporal trend modeling, providing higher accuracy and adaptability for STLF in complex power grid environments.

Enhanced Forecasting of Wind Energy Production: A Hybrid BPNN-SVR Model for Short-Term Wind Power Forecasting

Short-term photovoltaic power prediction based on a VMD-SE-CNN-LSTM combined model

To enhance power dispatching and mitigate grid connection fluctuations, this paper proposes a wind power prediction model based on Long Short-Term Memory-Back Propagation Neural Network (LSTM-BP) optimized by an adaptive Particle Swarm Optimization algorithm (aPSO). Initially, anomalies and missing values in raw wind farm data are addressed using the quartile method and filled via cubic spline interpolation. The data is then denoised using the Autoregressive Integrated Moving Average (ARIMA) model. Statistical and combined features are extracted, and Bayesian optimization is applied for optimal feature selection. To overcome the limitations of single models, a hybrid approach is adopted where a BP neural network is used in conjunction with LSTM. The optimal features are first input into the BP neural network to learn the current relationship between features and wind power. Then, historical data of both the features and wind power are fed into the LSTM to generate preliminary predictions. These LSTM outputs are subsequently passed into the trained BP neural network, and the final wind power prediction result is obtained through network integration. This combined model leverages the temporal learning capabilities of LSTM and the fitting strengths of BP, while aPSO ensures optimal parameter tuning, ultimately enhancing prediction accuracy and robustness in wind power forecasting. The experimental results show that the proposed model achieves a MAE of 0.54 MW and a MAPE of 10.5% in one-step prediction, reducing the error by over 35% compared to benchmark models such as ARIMA-LSTM and LSTM-BP. Multi-step prediction validation on 2000 sets of real wind farm data demonstrates the robustness and generalization capabilities of the proposed model.

ABSTRACT Wind power has achieved widespread application in electricity generation, and its installed capacity has demonstrated consistent annual growth. However, the inherent unpredictability and stochastic nature of wind resources result in unstable power output, which compromises grid reliability and operational security. To address these challenges, a lightweight network, Frequency-Enhanced Hierarchical Attention Network, has been developed. Specifically, the network captures frequency-domain features and spatial correlations in wind power data through its hierarchical attention module, and the Bidirectional Long Short-term Memory module enables robust modeling of long-term temporal dependencies. First, hierarchical attention incorporates frequency-domain information into both channel and spatial attention mechanisms through the Discrete Cosine Transform, effectively integrating spectral features and suppressing high-frequency noise. Furthermore, the integration of dual-attention mechanisms enables adaptive focus on critical features and spatiotemporal patterns via dynamic weight allocation. Second, the Bidirectional Long Short-term Memory enhances predictive capability for complex wind power data through bidirectional information flow and gated control mechanisms. Our result has been validated on two public datasets from Kaggle and a real-world dataset from power plants. Multiple comparative experiments demonstrate that the proposed network achieves higher predictive accuracy and has a lower computational training time.

This paper presents an optimized hybrid deep learning model for power load forecasting—QR-FMD-CNN-BiGRU-Attention—that integrates similar day selection, load decomposition, and deep learning to address the nonlinearity and volatility of power load data. Firstly, the original data are classified using Gaussian Mixture Clustering optimized by ICPO (ICPO-GMM), and similar day samples consistent with the predicted day category are selected. Secondly, the load data are decomposed into multi-scale components (IMFs) using feature mode decomposition optimized by ICPO (ICPO-FMD). Then, with the IMFs as targets, the quantile interval forecasting is trained using the CNN-BiGRU-Attention model optimized by ICPO. Subsequently, the forecasting model is applied to the features of the predicted day to generate interval forecasting results. Finally, the model’s performance is validated through comparative evaluation metrics, sensitivity analysis, and interpretability analysis. The experimental results show that compared with the comparative algorithm presented in this paper, the improved model has improved RMSE by at least 39.84%, MAE by 26.12%, MAPE by 45.28%, PICP and MPIW indicators by at least 3.80% and 2.27%, indicating that the model not only outperforms the comparative model in accuracy, but also exhibits stronger adaptability and robustness in complex load fluctuation scenarios.

ABSTRACT The utilization of solar energy plays a significant role in alleviating the energy crisis and promoting sustainable development. However, photovoltaic (PV) power generation is impacted by several complex factors that cause discontinuity and randomness problems. This paper presents a novel approach to improving the precision of PV power forecasting, based on a LASSO-penalized multi-core learning-based robust regression algorithm with random forest (RF) optimization. This was further enhanced using particle swarm optimization (PSO). Initially, the PSO-RF was employed to identify the key variables influencing the PV output power. Second, the issue of a shallow Gaussian multi-core structure was addressed by employing the multi-kernel learning (MKL) method, which was used to construct a highly robust depth mapping kernel based on the multi-layer information of deep neural networks. A multiscale Gaussian kernel was combined with a kernel matrix to create an improved kernel. This paper presents a novel computational approach based on local quadratic approximation for solving models. The findings indicate that the robust multi-core sparse prediction model developed in this paper achieves enhanced prediction accuracy and robustness. Regarding the coefficient of determination, the PSORF-LPMKLRRA model achieves the highest value of 0.998, demonstrating a closer approximation to 1 than the alternative predictive models.

Joint application of Crested Porcupine Optimizer and hybrid models in short-term wind power load forecasting

With the rapid growth of the global renewable energy industry, photovoltaic (PV) power generation has become a crucial component of clean energy systems. However, PV power output exhibits significant variability and uncertainty due to weather conditions, seasonal shifts, and geographical differences. Developing an accurate and robust PV power prediction model is essential for improving the operational efficiency of PV plants, reducing power system regulation costs, and increasing the utilization rate of renewable energy. This study proposes a novel PV power forecasting model based on the State-Frequency Memory (SFM) recurrent network. The model incorporates a time-varying transformation mechanism for specific sequences, combined with discrete Fourier transform-based frequency decomposition and weight adjustment techniques, to effectively handle the frequency information in PV power output, thereby enhancing prediction accuracy. To evaluate the model’s effectiveness, historical datasets from the National Renewable Energy Laboratory in the United States were used for both short-term and seasonal PV forecasting. Comparative experiments with conventional short-term prediction models demonstrate that the proposed SFM-based model achieves superior performance, with the lowest root mean square error and mean absolute error values across various time scales and seasonal conditions. These results confirm the model’s high accuracy, robustness, and practical value in supporting reliable PV power forecasting.

Short-term wind power forecast with turning weather based on DBSCAN-RFE-LightGBM