2026· E3S Web of Conferences· Vol 720, pp. 02002· 0 citations· 12 references
TL;DR
The proposed DGA-DANN-TCN-LSTM framework realizes the high accuracy prediction of multivariate meteorological data across time domains, and provides an effective method for water-wind-photovoltaic complementary scheduling.
Abstract
Accurate meteorological forecasts are crucial for optimizing wind-solar scheduling systems, but increasing climate anomalies have led to significant changes in the statistical distribution of meteorological data over time. Conventional deep learning models struggle with these non-stationary distributions. To address this, we propose the DGA-DANN-TCN-LSTM framework. This domain-adaptive approach incorporates a domain-guided attention (DGA) mechanism, utilizing a global mean query to align source and target feature distributions through adversarial training to effectively capture overall climate contexts.Using ERA5 reanalysis data for the Jinsha River basin, experiments demonstrate the proposed model significantly outperforms the TCN-LSTM and DANN-TCN-LSTM. Under extremely sparse data conditions with only 30 days of training data, the DGA mechanism successfully captured seasonal patterns. Compared to the TCN-LSTM model, wind speed prediction R² improved by 91.8%, and the root mean square error decreased by 49.7%. By compensating for the standard DANN model’s neglect of global seasonal trends, the proposed model achieves prediction accuracy comparable to the standard DANN-TCN-LSTM model trained on 365 days of data when trained with only 90 days of data. Under full-year data conditions, this model still reduces the root mean square error by 10% to 33% compared to the TCN-LSTM model. The proposed model realizes the high accuracy prediction of multivariate meteorological data across time domains, and provides an effective method for water-wind-photovoltaic complementary scheduling.
Predictions of solar power output that are close to the mark may help with grid stability, energy management, and the integration of renewable energy sources. Prediction using conventional statistical models is challenging due to the substantial influence of nonlinear meteorological variables and temporal variations on solar power output. A deep learning-based forecasting framework integrating an attention mechanism and a Temporal Convolutional Network (TCN) is proposed in this study to enhance the precision of solar energy prediction. A number of feature engineering and pre-processing approaches are used by the model, including lag features, rolling statistics, temporal indicators, and outlier reduction using the Interquartile Range (IQR) methodology. To store long-range temporal associations, the TCN architecture employs dilated causal convolutions; the attention mechanism highlights the most important time steps in the input sequence. The proposed model outperforms state-of-the-art baseline models such as Long Short-Term Memory (LSTM) and traditional TCN when tested on a publicly available solar energy dataset. It can accomplish 253.32 MW Mean Absolute Error (MAE), 325.04 MW Root Mean Square Error (RMSE), 2.30% Mean Absolute Percentage Error (MAPE), and 0.9804 Coefficient of Determination (R2). These parameters are evaluated by experimental means.
Mohamed Shaik Honnurvali, Mazhar Baloch, Touqeer Ahmed et al.· IEEE Open Access Journal of...· 0 citations
Short-horizon forecasting of fine particulate matter (PM2.5) remains difficult when observations from the target domain are limited and the statistical properties of the source and target domains differ. In these settings, models trained only on local data may not capture complex temporal dynamics, while direct transfer learning can result in negative transfer. This study develops a shift-aware dual-encoder transfer framework that combines source-domain knowledge with target-specific representation learning. The source encoder was pretrained using hourly observations from 10 U.S. monitoring locations. The framework was then adapted and evaluated using two years of hourly observations from 77 stations in Taiwan under a chronological train-validation-test protocol. Among the four principal baselines, the frozen-source dual-encoder model achieved the best performance, with MSE = 21.8960, MAE = 3.1597, and R^2 = 0.8725. This corresponds to an MSE reduction of approximately 7.1% relative to TL-v1 and 4.1% relative to TL-v2. The ablation analysis showed that removing the Taiwan-specific branch caused the largest decline in performance. Allowing the source encoder to adapt produced the best overall result, with MSE = 21.6575, MAE = 3.1383, and R^2 = 0.8739. SHAP analysis indicated that predictions were driven mainly by recent PM2.5 observations and meteorological variables related to pollutant transport and dispersion. These results suggest that source-domain knowledge is most effective when target-specific information is preserved and the transferred representation is allowed to adapt under target supervision.
The findings highlight the value of training strategies that allow models to directly learn bias correction during forecast transitions, emphasize the operational potential of combining sequential processing with near real-time discharge observations and identify physiographic catchment characteristics as key modulators of forecast skill improvement across diverse hydroclimatic settings.
O. Konold, Moritz Feigl, Patrick Podest et al.· Hydrology and Earth System S...· 2 citations
Accurate wind speed downscaling is essential for meteorological applications and reliable station-scale wind estimation. This study presents a systematic comparison of recurrent deep learning architectures, including RNN, GRU, LSTM, Bidirectional LSTM, Sequence-to-Sequence LSTM, and Stacked LSTM, against conventional statistical models for wind speed downscaling using paired IMD station observations and NASA MERRA-2 reanalysis data for Pune, India. All models were evaluated under identical preprocessing and training protocols to ensure a fair comparison. Among the standard recurrent architectures, the Stacked LSTM achieved the best predictive performance. An Attention-Enhanced LSTM was subsequently evaluated as an extension to investigate the benefits of incorporating a soft-attention mechanism. It achieved the highest overall accuracy, with an RMSE of 0.5034 and an MAE of 0.3650, demonstrating additional performance gains over the standard recurrent architectures. The findings provide a robust benchmark for deep learning-based wind speed downscaling and future atmospheric prediction studies.
Juhi Kumari, Rajesh Wadhvani, Sanyam Shukla· Wind Engineering : The Inter...· 0 citations
A new hierarchical forecasting structure denoted as INRBO-SSA-LSTM, which significantly outperforms traditional baseline models across all indicators and effectively accommodates temporal demand variations, offering a robust foundation for the advancement of intelligent power management technology.
Jinming Luo, Fu Chen, Lingshan Kong et al.· Electronics· 0 citations
Deep learning’s significant potential as a scalable and dependable method for regional climate forecasting, giving it an advantage in the Kathmandu Valley’s climate change adaptation policy-making process is supported.
Bikash Chawal, K. Poudyal, Mahammad Humayoo· Journal of Sciences and Engi...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
