A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model

Abstract Photovoltaic (PV) power is significantly influenced by meteorological fluctuations, and its forecasting accuracy is critical for power system dispatching and economic operation. To enhance forecasting precision, this paper proposes a hybrid framework integrating signal decomposition, parall...

Full description

Saved in:
Bibliographic Details
Main Authors: Ying Xu, Xinrong Ji, Zhengyang Zhu
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-16368-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849233995644010496
author Ying Xu
Xinrong Ji
Zhengyang Zhu
author_facet Ying Xu
Xinrong Ji
Zhengyang Zhu
author_sort Ying Xu
collection DOAJ
description Abstract Photovoltaic (PV) power is significantly influenced by meteorological fluctuations, and its forecasting accuracy is critical for power system dispatching and economic operation. To enhance forecasting precision, this paper proposes a hybrid framework integrating signal decomposition, parallel forecasting, and weight optimization. Firstly, the Thompson-Tau-Newton interpolation method is applied to handle missing data, and key meteorological factors are selected using the Pearson correlation coefficient to reduce input dimensionality. Secondly, the power sequence is decomposed into multi-scale subsequences using Ensemble Empirical Mode Decomposition (EEMD), which are then reconstructed into low-frequency components (reflecting trend features) and high-frequency components (capturing random fluctuations) based on sample entropy. Furthermore, a parallel XGBoost-LSTM forecasting structure is constructed, XGBoost models the low-frequency components to capture global patterns, while LSTM processes the high-frequency components to learn temporal dependencies. Finally, the Snake Optimization (SO) algorithm is introduced to dynamically optimize the combination weights, enabling adaptive fusion of forecasting results. Experimental results demonstrate that the proposed model significantly outperforms standalone benchmark methods. In comparison with Particle Swarm Optimization (PSO), Sparrow Search Algorithm (SSA), and the equal-weight assignment approach for high- and low-frequency component forecasting, the proposed SO algorithm attains the lowest forecasting errors. The proposed method provides a novel approach to high-precision PV power forecasting by integrating multi-modal feature fusion and optimized weight allocation.
format Article
id doaj-art-0c138d58640f49f2bb1d88ce24a49e0a
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-0c138d58640f49f2bb1d88ce24a49e0a2025-08-20T04:03:18ZengNature PortfolioScientific Reports2045-23222025-08-0115111310.1038/s41598-025-16368-9A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO modelYing Xu0Xinrong Ji1Zhengyang Zhu2State Grid Wuxi Power Supply CompanyState Grid Jiangsu Electric Power Co., LtdState Grid Nantong Power Supply CompanyAbstract Photovoltaic (PV) power is significantly influenced by meteorological fluctuations, and its forecasting accuracy is critical for power system dispatching and economic operation. To enhance forecasting precision, this paper proposes a hybrid framework integrating signal decomposition, parallel forecasting, and weight optimization. Firstly, the Thompson-Tau-Newton interpolation method is applied to handle missing data, and key meteorological factors are selected using the Pearson correlation coefficient to reduce input dimensionality. Secondly, the power sequence is decomposed into multi-scale subsequences using Ensemble Empirical Mode Decomposition (EEMD), which are then reconstructed into low-frequency components (reflecting trend features) and high-frequency components (capturing random fluctuations) based on sample entropy. Furthermore, a parallel XGBoost-LSTM forecasting structure is constructed, XGBoost models the low-frequency components to capture global patterns, while LSTM processes the high-frequency components to learn temporal dependencies. Finally, the Snake Optimization (SO) algorithm is introduced to dynamically optimize the combination weights, enabling adaptive fusion of forecasting results. Experimental results demonstrate that the proposed model significantly outperforms standalone benchmark methods. In comparison with Particle Swarm Optimization (PSO), Sparrow Search Algorithm (SSA), and the equal-weight assignment approach for high- and low-frequency component forecasting, the proposed SO algorithm attains the lowest forecasting errors. The proposed method provides a novel approach to high-precision PV power forecasting by integrating multi-modal feature fusion and optimized weight allocation.https://doi.org/10.1038/s41598-025-16368-9Photovoltaic power forecastingEnsemble empirical mode decomposition (EEMD)Long Short-Term memory (LSTM)XGBoostSnake optimization (SO)Hybrid model
spellingShingle Ying Xu
Xinrong Ji
Zhengyang Zhu
A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
Scientific Reports
Photovoltaic power forecasting
Ensemble empirical mode decomposition (EEMD)
Long Short-Term memory (LSTM)
XGBoost
Snake optimization (SO)
Hybrid model
title A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
title_full A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
title_fullStr A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
title_full_unstemmed A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
title_short A photovoltaic power forecasting method based on the LSTM-XGBoost-EEDA-SO model
title_sort photovoltaic power forecasting method based on the lstm xgboost eeda so model
topic Photovoltaic power forecasting
Ensemble empirical mode decomposition (EEMD)
Long Short-Term memory (LSTM)
XGBoost
Snake optimization (SO)
Hybrid model
url https://doi.org/10.1038/s41598-025-16368-9
work_keys_str_mv AT yingxu aphotovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel
AT xinrongji aphotovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel
AT zhengyangzhu aphotovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel
AT yingxu photovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel
AT xinrongji photovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel
AT zhengyangzhu photovoltaicpowerforecastingmethodbasedonthelstmxgboosteedasomodel