Leveraging advanced AI algorithms with transformer-infused recurrent neural networks to optimize solar irradiance forecasting DOI Creative Commons

Muhammad Sabir Naveed,

Muhammad Fainan Hanif, Mohamed Metwaly

et al.

Frontiers in Energy Research, Journal Year: 2024, Volume and Issue: 12

Published: Oct. 8, 2024

Solar energy (SE) is vital for renewable generation, but its natural fluctuations present difficulties in maintaining grid stability and planning. Accurate forecasting of solar irradiance (SI) essential to address these challenges. The current research presents an innovative approach named as Transformer-Infused Recurrent Neural Network (TIR) model. This model integrates a Bi-Directional Long Short-Term Memory (BiLSTM) network encoding Gated Unit (GRU) decoding, incorporating attention mechanisms positional encoding. proposed enhance SI accuracy by effectively utilizing meteorological weather data, handling overfitting, managing data outliers complexity. To evaluate the model’s performance, comprehensive comparative analysis conducted, involving five algorithms: Artificial (ANN), BiLSTM, GRU, hybrid BiLSTM-GRU, Transformer models. findings indicate that employing TIR leads superior analyzed area, achieving R 2 value 0.9983, RMSE 0.0140, MAE 0.0092. performance surpasses those alternative models studied. integration BiLSTM GRU algorithms with mechanism has been optimized SI. mitigates computational dependencies minimizes error terms within

Language: Английский

Enhancing Solar Forecasting Accuracy with Sequential Deep Artificial Neural Network and Hybrid Random Forest and Gradient Boosting Models across Varied Terrains DOI
Muhammad Farhan Hanif,

Muhammad Umar Siddique,

Jicang Si

et al.

Advanced Theory and Simulations, Journal Year: 2024, Volume and Issue: 7(7)

Published: April 30, 2024

Abstract Effective solar energy utilization demands improvements in forecasting due to the unpredictable nature of irradiance (SI). This study introduces and rigorously tests two innovative models across different locations: Sequential Deep Artificial Neural Network (SDANN) Hybrid Random Forest Gradient Boosting (RFGB). SDANN, leveraging deep learning, aims identify complex patterns weather data, while RFGB, combining Boosting, proves more effective by offering a superior balance efficiency accuracy. The research highlights SDANN model's learning capabilities along with RFGB unique blend their comparative success over existing such as eXtreme (XGBOOST), Categorical (CatBOOST), Gated Recurrent Unit (GRU), K‐Nearest Neighbors (KNN) XGBOOST hybrid. With lowest Mean Squared Error (147.22), Absolute (8.77), high R 2 value (0.80) studied region, stands out. Additionally, detailed ablation studies on meteorological feature impacts model performance further enhance accuracy adaptability. By integrating cutting‐edge AI SI forecasting, this not only advances field but also sets stage for future renewable strategies global policy‐making.

Language: Английский

Citations

5
AI-Driven precision in solar forecasting: Breakthroughs in machine learning and deep learning DOI Creative Commons
Ayesha Nadeem, Muhammad Farhan Hanif,

Muhammad Sabir Naveed

et al.

AIMS Geosciences, Journal Year: 2024, Volume and Issue: 10(4), P. 684 - 734

Published: Jan. 1, 2024

<p>The need for accurate solar energy forecasting is paramount as the global push towards renewable intensifies. We aimed to provide a comprehensive analysis of latest advancements in forecasting, focusing on Machine Learning (ML) and Deep (DL) techniques. The novelty this review lies its detailed examination ML DL models, highlighting their ability handle complex nonlinear patterns Solar Irradiance (SI) data. systematically explored evolution from traditional empirical, including machine learning (ML), physical approaches these advanced delved into real-world applications, discussing economic policy implications. Additionally, we covered variety image-based, statistical, ML, DL, foundation, hybrid models. Our revealed that models significantly enhance accuracy, operational efficiency, grid reliability, contributing benefits supporting sustainable policies. By addressing challenges related data quality model interpretability, underscores importance continuous innovation techniques fully realize potential. findings suggest integrating with offers most promising path forward improving forecasting.</p>

Language: Английский

Citations

5
Enhanced accuracy in solar irradiance forecasting through machine learning stack-based ensemble approach DOI

Muhammad Sabir Naveed,

Hafiz M.N. Iqbal, Muhammad Fainan Hanif

et al.

International Journal of Green Energy, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 24

Published: Jan. 10, 2025

Accurate solar irradiance (SI) prediction is vital for optimizing photovoltaic systems. This study addresses shortcomings in existing forecasting methods by exploring advanced machine-learning techniques using meteorological satellite data. We develop three novel models SI forecasting: Stack-based Ensemble Fusion with Meta-Neural Network (SEFMNN), Extreme Gradient Boosting-Squared Error (XGB-SE), and Learning Machine (ELM). These predict All-sky Clear-sky shortwave across Chinese provinces (Guangdong, Shandong, Zhejiang) one Saudi Arabian province (Najran). The SEFMNN model combines Artificial Neural (ANN), Random Forest (RF), Support Vector (SVM) to improve accuracy. XGB-SE employs a specialized loss function manage extreme values historical are designed mitigate overfitting data inconsistency while balancing computational efficiency predictive Comparative analysis reveals that outperform the ELM model, achieving an R2 of 0.9979, MAE 0.0231, MSE 0.0020 Najran. demonstrates significantly enhances forecasting, aiding efficient system planning operation.

Language: Английский

Citations

0
Explainable deep learning hybrid modeling framework for total suspended particles concentrations prediction DOI
Sujan Ghimire, Ravinesh C. Deo, Ningbo Jiang

et al.

Atmospheric Environment, Journal Year: 2025, Volume and Issue: unknown, P. 121079 - 121079

Published: Feb. 1, 2025

Language: Английский

Citations

0
Enhancing solar radiation forecasting accuracy with a hybrid SA-Bi-LSTM-Bi-GRU model DOI

Girijapati Sharma,

Subhash Chandra,

Arvind Yadav

et al.

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(3)

Published: Feb. 19, 2025

Language: Английский

Citations

0
Ensemble-Empirical-Mode-Decomposition (EEMD) on SWH prediction: The effect of decomposed IMFs, continuous prediction duration, and data-driven models DOI

Yuanye Guo,

Jicang Si, Yulian Wang

et al.

Ocean Engineering, Journal Year: 2025, Volume and Issue: 324, P. 120755 - 120755

Published: Feb. 24, 2025

Language: Английский

Citations

0
Leveraging advanced AI algorithms with transformer-infused recurrent neural networks to optimize solar irradiance forecasting DOI Creative Commons

Muhammad Sabir Naveed,

Muhammad Fainan Hanif, Mohamed Metwaly

et al.

Frontiers in Energy Research, Journal Year: 2024, Volume and Issue: 12

Published: Oct. 8, 2024

Solar energy (SE) is vital for renewable generation, but its natural fluctuations present difficulties in maintaining grid stability and planning. Accurate forecasting of solar irradiance (SI) essential to address these challenges. The current research presents an innovative approach named as Transformer-Infused Recurrent Neural Network (TIR) model. This model integrates a Bi-Directional Long Short-Term Memory (BiLSTM) network encoding Gated Unit (GRU) decoding, incorporating attention mechanisms positional encoding. proposed enhance SI accuracy by effectively utilizing meteorological weather data, handling overfitting, managing data outliers complexity. To evaluate the model’s performance, comprehensive comparative analysis conducted, involving five algorithms: Artificial (ANN), BiLSTM, GRU, hybrid BiLSTM-GRU, Transformer models. findings indicate that employing TIR leads superior analyzed area, achieving R 2 value 0.9983, RMSE 0.0140, MAE 0.0092. performance surpasses those alternative models studied. integration BiLSTM GRU algorithms with mechanism has been optimized SI. mitigates computational dependencies minimizes error terms within

Language: Английский

Citations

1