Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review

Accurate solar and photovoltaic (PV) power forecasting is essential for optimizing grid integration, managing energy storage, and maximizing the efficiency of solar power systems. Deep learning (DL) models have shown promise in this area due to their ability to learn complex, non-linear relationship...

Full description

Saved in:
Bibliographic Details
Main Authors: Oussama Khouili, Mohamed Hanine, Mohamed Louzazni, Miguel Angel López Flores, Eduardo García Villena, Imran Ashraf
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Energy Strategy Reviews
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2211467X25000987
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849724677003411456
author Oussama Khouili
Mohamed Hanine
Mohamed Louzazni
Miguel Angel López Flores
Eduardo García Villena
Imran Ashraf
author_facet Oussama Khouili
Mohamed Hanine
Mohamed Louzazni
Miguel Angel López Flores
Eduardo García Villena
Imran Ashraf
author_sort Oussama Khouili
collection DOAJ
description Accurate solar and photovoltaic (PV) power forecasting is essential for optimizing grid integration, managing energy storage, and maximizing the efficiency of solar power systems. Deep learning (DL) models have shown promise in this area due to their ability to learn complex, non-linear relationships within large datasets. This study presents a systematic literature review (SLR) of deep learning applications for solar PV forecasting, addressing a gap in the existing literature, which often focuses on traditional ML or broader renewable energy applications. This review specifically aims to identify the DL architectures employed, preprocessing and feature engineering techniques used, the input features leveraged, evaluation metrics applied, and the persistent challenges in this field. Through a rigorous analysis of 26 selected papers from an initial set of 155 articles retrieved from the Web of Science database, we found that Long Short-Term Memory (LSTM) networks were the most frequently used algorithm (appearing in 32.69% of the papers), closely followed by Convolutional Neural Networks (CNNs) at 28.85%. Furthermore, Wavelet Transform (WT) was found to be the most prominent data decomposition technique, while Pearson Correlation was the most used for feature selection. We also found that ambient temperature, pressure, and humidity are the most common input features. Our systematic evaluation provides critical insights into state-of-the-art DL-based solar forecasting and identifies key areas for upcoming research. Future research should prioritize the development of more robust and interpretable models, as well as explore the integration of multi-source data to further enhance forecasting accuracy. Such advancements are crucial for the effective integration of solar energy into future power grids.
format Article
id doaj-art-b2008cf42f6b481d8ecc332f9dce74d9
institution DOAJ
issn 2211-467X
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series Energy Strategy Reviews
spelling doaj-art-b2008cf42f6b481d8ecc332f9dce74d92025-08-20T03:10:41ZengElsevierEnergy Strategy Reviews2211-467X2025-05-015910173510.1016/j.esr.2025.101735Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic reviewOussama Khouili0Mohamed Hanine1Mohamed Louzazni2Miguel Angel López Flores3Eduardo García Villena4Imran Ashraf5LTI Laboratory, National School of Applied Sciences, Chouaib Doukkali University, El Jadida, 24000, MoroccoLTI Laboratory, National School of Applied Sciences, Chouaib Doukkali University, El Jadida, 24000, MoroccoScience Engineer Laboratory for Energy, National School of Applied Sciences, Chouaib Doukkali University of El Jadida, MoroccoUniversidad Europea del Atlántico, Isabel Torres 21, 39011, Santander, Spain; Universidad Internacional Iberoamericana Campeche, 24560, Mexico; Instituto Politécnico Nacional, UPIICSA, Ciudad de México, MexicoUniversidad Europea del Atlántico, Isabel Torres 21, 39011, Santander, Spain; Universidad Internacional Iberoamericana Arecibo, Puerto Rico, 00613, USA; Universidade Internacional do Cuanza, Cuito Bié, AngolaInformation and Communication Engineering, Yeungnam University, Gyeongsan, 38541, Republic of Korea; Corresponding author.Accurate solar and photovoltaic (PV) power forecasting is essential for optimizing grid integration, managing energy storage, and maximizing the efficiency of solar power systems. Deep learning (DL) models have shown promise in this area due to their ability to learn complex, non-linear relationships within large datasets. This study presents a systematic literature review (SLR) of deep learning applications for solar PV forecasting, addressing a gap in the existing literature, which often focuses on traditional ML or broader renewable energy applications. This review specifically aims to identify the DL architectures employed, preprocessing and feature engineering techniques used, the input features leveraged, evaluation metrics applied, and the persistent challenges in this field. Through a rigorous analysis of 26 selected papers from an initial set of 155 articles retrieved from the Web of Science database, we found that Long Short-Term Memory (LSTM) networks were the most frequently used algorithm (appearing in 32.69% of the papers), closely followed by Convolutional Neural Networks (CNNs) at 28.85%. Furthermore, Wavelet Transform (WT) was found to be the most prominent data decomposition technique, while Pearson Correlation was the most used for feature selection. We also found that ambient temperature, pressure, and humidity are the most common input features. Our systematic evaluation provides critical insights into state-of-the-art DL-based solar forecasting and identifies key areas for upcoming research. Future research should prioritize the development of more robust and interpretable models, as well as explore the integration of multi-source data to further enhance forecasting accuracy. Such advancements are crucial for the effective integration of solar energy into future power grids.http://www.sciencedirect.com/science/article/pii/S2211467X25000987Deep learningPV power forecastingSolar radiation forecastingSystematic review
spellingShingle Oussama Khouili
Mohamed Hanine
Mohamed Louzazni
Miguel Angel López Flores
Eduardo García Villena
Imran Ashraf
Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
Energy Strategy Reviews
Deep learning
PV power forecasting
Solar radiation forecasting
Systematic review
title Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
title_full Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
title_fullStr Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
title_full_unstemmed Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
title_short Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review
title_sort evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting a systematic review
topic Deep learning
PV power forecasting
Solar radiation forecasting
Systematic review
url http://www.sciencedirect.com/science/article/pii/S2211467X25000987
work_keys_str_mv AT oussamakhouili evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview
AT mohamedhanine evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview
AT mohamedlouzazni evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview
AT miguelangellopezflores evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview
AT eduardogarciavillena evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview
AT imranashraf evaluatingtheimpactofdeeplearningapproachesonsolarandphotovoltaicpowerforecastingasystematicreview