What type of study is this?

September 5, 2025Open Access

Comparative analysis of deep learning architectures in solar power prediction

MAMontaser AbdelsattarSouth Valley University MAMohamed Mostafa A. AzimPrince Sultan University AAAhmed AbdelMoetySouth Valley University

Key Points

The temporal convolutional network outperforms other models, achieving an R² of 0.7786 for solar power predictions.
With an overall R² of 0.8437, the autoencoder demonstrates strong performance across the entire dataset.
The study assesses deep learning architectures using metrics like RMSE and MAE on historical solar data.
Findings highlight the importance of integrating reliable deep learning models into smart grid systems for effective forecasting.

Abstract

Abstract Integrating renewable energy sources into the electricity grid requires accurate forecasts of solar power production. With the aim of enhancing the accuracy and reliability of forecasts, this study presents a comprehensive comparative analysis of eight state-of-the-art Deep Learning (DL) architectures—Autoencoder, Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Simple Recurrent Neural Network (SimpleRNN), Convolutional Neural Network (CNN), Temporal Convolutional Network (TCN), Transformer, and Lightweight Informer for Long Sequence Time-Series Forecasting (InformerLite) —applied to solar power prediction using a dataset with 4, 200 historical records and 20 meteorological and astronomical features. A comprehensive assessment of Root Mean Squared Error \: (RMSE), Mean Absolute Error \: (MAE), Mean Absolute Percentage Error \: (MAPE), and Coefficient of Determination \: (R^2) metrics was performed on the training, validation, and test datasets. The TCN model had the greatest performance across all models, achieving a test R² of 0. 7786, an \: RMSE of 429. 4863, and a balanced relative standard deviation (\: RSD) of 0. 6827, so exhibiting an exceptional capacity to capture temporal patterns. The Autoencoder achieved a \: R^2 of 0. 7648 and had the greatest overall performance on the entire dataset, resulting in a Whole \: R^2 of 0. 8437. In contrast, the Transformer model demonstrated significantly poorer performance (Test \: R^2 = 0. 0714), underscoring its limitations in this context without any architectural modifications. This study not only demonstrates the best DL models for solar power forecasting as qualified by useful statistical metrics, but also provides a scalable, interpretable, and extensible forecasting framework for real-world energy systems. The findings verify the informed DL integration to smart grid scenarios, laying the foundations for further developments in hybrid modeling, multi-horizon prediction, and deployment in resource-constrained environments with limited computational power and resources.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper