Aug 2026· Energy Exploration & Exploitation· 0 citations· 49 references
TL;DR
The integration of physics-guided constraints with the temporal convolutional architecture significantly enhances prediction accuracy, stability, and generalization capability, making it suitable for real-time wind energy forecasting applications, intelligent energy management systems, and microgrid power system operations.
Abstract
Accurate short-term wind power prediction plays a critical role in ensuring stable grid operation, effective energy management, and the large-scale integration of renewable energy systems under highly variable wind conditions. Although data-driven and deep learning models have demonstrated promising forecasting capability, many existing approaches suffer from performance degradation during rapid wind fluctuations due to the lack of embedded physical constraints and the high computational complexity associated with recurrent architectures. To address these limitations, this article proposes a Physics-Guided Residual Temporal Convolutional Network (PG-ResTCN) for short-term wind power forecasting. The proposed framework integrates dilated temporal convolutional learning with residual connections to effectively capture multiscale temporal dependencies in wind power time series while avoiding the sequential computation of recurrent neural networks, thereby improving computational efficiency. Furthermore, a physics-based smoothness constraint is incorporated into the training loss function to enforce physically consistent power ramping behavior that reflects the inherent operational dynamics and inertia of wind turbines, reducing unrealistic fluctuations in predicted power outputs. The effectiveness of the proposed model is validated using a large real-world dataset containing approximately 140,161 hourly samples collected from five wind farm locations, including meteorological variables and corresponding turbine power outputs. Comprehensive experiments are conducted by comparing the proposed method with widely used benchmark models, including Support Vector Regression, Random Forest, Extreme Gradient Boosting, and Long Short-Term Memory networks. Results demonstrate that the proposed PG-ResTCN model achieves superior forecasting performance, obtaining a root mean square error of 0.0359, mean absolute error of 0.0296, and R
2
of 0.9759, outperforming all baseline models. The integration of physics-guided constraints with the temporal convolutional architecture significantly enhances prediction accuracy, stability, and generalization capability. In addition, the proposed framework maintains high computational efficiency and robustness, making it suitable for real-time wind energy forecasting applications, intelligent energy management systems, and microgrid power system operations.
Given the problems of non-stationary power time series, response lag, and amplified prediction errors due to sudden changes in wind direction under transitional meteorological conditions, this study proposes a wind power forecasting model that integrates multiscale time-series features, transition-aware attention mechanisms, physical constraints on turbine operation, and dynamic residual correction. To improve the model's ability to jointly characterize different scales of meteorological evolution and power lag characteristics, a feature extraction network with short-, medium-, and long-term branches based on TCN (Temporal Convolutional Network) is built, which incorporates bidirectional GRU and Transformer architectures; additionally, the feature weights at each scale are dynamically adjusted according to the severity of inflection points, and the prediction output is constrained by air density, power curves, operational status and ramping limits. Residual caching and time-delay gating are employed to correct for lag errors in the range of 1 to 6 sampling steps. 50,842 valid data sets from a wind farm were used for validation. The general MAE and RMSE of the model were 0.258 MW and 0.386 MW, respectively, and these were lower than those of the baseline TCN by 21.58% and 20.90%. In the transition period, the MAE and RMSE were 0.305 MW and 0.448 MW; these had been reduced by 22.19% and 23.29%. Therefore, the developed model can reduce peak deviations caused by abrupt changes in weather and improve the accuracy and stability of wind power forecasting under all operating conditions.
Y.-J. Li, J. Shen, S. Xu et al.· Advanced Electromagnetics· 0 citations
With the large‐scale integration of wind power, its inherent intermittency and uncertainty pose significant challenges to power system operation. Existing probabilistic forecasting methods often struggle with capturing spatiotemporal correlations among multiple turbines as well as generalizing to unseen turbines or future time steps. To address these issues, we propose a novel multi‐turbine collaborative probabilistic forecasting framework based on the flow matching enhanced neural process. By formulating wind power forecasting as a conditional stochastic process, our method naturally enables cross‐turbine knowledge sharing through a Transformer‐based encoder. The introduction of flow matching allows efficient single‐ or few‐step sampling while avoiding the usage of variational lower bounds, and hence is more effective. Experiments on a real‐world dataset comprising 134 turbines over 245 days demonstrate that the proposed framework achieves superior performance, significantly outperforming existing neural process variants and deep learning baselines in terms of RMSE, MAE, and continuous ranked probability score.
Jiabei Liu, Zheren Zhu, Le Yao et al.· International Journal of Ada...· 0 citations
A novel hybrid intelligent framework—integrating Improved Lotus Effect Algorithm, Variational Mode Decomposition, and ensemble deep learning—specifically designed for ultra-short-term wind power prediction in energy dispatch applications is engineers, which employs elite chaotic opposition-based learning to autonomously optimize VMD decomposition levels.
Lei Shen, Qifeng Xiang, Q. Gao et al.· Energy Engineering· 0 citations
Accurate wind power forecasting is essential for renewable-energy accommodation, low-carbon dispatch, and the sustainable operation of modern power systems. However, wind power series exhibit pronounced non-stationarity, strong volatility, and multi-scale evolution, making long-term trends and short-term disturbances difficult to characterize jointly. In addition, multi-step forecasting errors tend to accumulate with increasing horizons, degrading model accuracy and stability. To address these issues, this study proposes CoFFormer, a collaborative frequency-domain-enhanced network for non-stationary wind power forecasting. The model reduces input modeling complexity, strengthens collaborative representation of heterogeneous temporal information, and suppresses output-stage error accumulation. Specifically, embedded series decomposition mitigates coupling interference between trend and fluctuation components. Differentiated temporal modeling and dynamic gating then adaptively coordinate the contributions of different feature representations, while frequency-domain residual compensation enhances the recovery of periodic structures and local oscillations. Experiments on ETTh2, wind_speed, WindPower, and Location2 demonstrate strong competitiveness across forecasting horizons. CoFFormer achieves MSE/MAE values of 0.0957/0.2238 and 0.1508/0.2889 on ETTh2 for 12- and 24-step forecasting, and 0.0617/0.1490 and 0.3838/0.3948 on WindPower for 3- and 24-step forecasting, outperforming most baselines. Ablation studies confirm the effectiveness and synergy of each component, providing an effective solution for high-accuracy multi-step forecasting of complex non-stationary wind power series.
Accurate wind forecasting is critical to ensure stable and efficient integration of renewable energy resources in modern power systems. However, the inherent variability and non-stationarity of wind pose a significant forecasting problem for modern power system operators to ensure power system stability. A new hybrid forecasting model incorporating a Transformer Encoder and XGBoost regressor is proposed in this paper to enhance one-hour-ahead short-term forecasting of wind power generation. The model employs Transformer Encoders to exploit complex temporal relationships from sequences of 24-hour past wind data, learning daily and periodic trends, ramps, and hidden dynamics. The generated temporal feature vectors are then combined with meteorological variables such as wind speed and direction at various altitudes and passed on to the XGBoost model to obtain regression forecasts. The proposed approach has been validated on real-world datasets obtained from commercial wind energy farms. Experimental results demonstrate that the hybrid model outperforms standalone Transformer and XGBoost models by achieving accurate values for mean squared error, mean absolute error and coefficient of determination. In addition, model has tested for long-term forecasting as well. Although the model is effective for short-term forecasting, due to increased uncertainty and error propagation, its performance degrades with extended forecasting horizons.
Heshan Senapriya, Sakun Rasilka, D. P. Wadduwage· Moratuwa Engineering Researc...· 0 citations
A hybrid FCM-WGM-BiLSTM-Transformer (FW-BTP) framework integrating Fuzzy C-Means clustering, Weighted Grey Model (WGM) trend extraction, and a coupled BiLSTM-Transformer module is proposed, supporting refined scheduling in modern power systems.
