A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework

Key factors influencing photovoltaic (PV) power generation predictions encompass solar radiation, aerosols, sunshine duration, temperature, humidity, wind direction, wind speed, cloud cover, and so on. The various influencing factors exhibit nonlinear correlation correlations, causing high volatilit...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengji Yang, Haiqing Zhang, Xi Yu, Aicha Sekhari Seklouli, Abdelaziz Bouras, Yacine Ouzrout
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:International Journal of Electrical Power & Energy Systems
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S014206152500211X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850116871545683968
author Mengji Yang
Haiqing Zhang
Xi Yu
Aicha Sekhari Seklouli
Abdelaziz Bouras
Yacine Ouzrout
author_facet Mengji Yang
Haiqing Zhang
Xi Yu
Aicha Sekhari Seklouli
Abdelaziz Bouras
Yacine Ouzrout
author_sort Mengji Yang
collection DOAJ
description Key factors influencing photovoltaic (PV) power generation predictions encompass solar radiation, aerosols, sunshine duration, temperature, humidity, wind direction, wind speed, cloud cover, and so on. The various influencing factors exhibit nonlinear correlation correlations, causing high volatility and discreteness in PV power time series. Firstly, to reduce the redundancy of the input for the prediction model and the computational time complexity, while enhancing the robustness and stability of the prediction model, nonlinear correlation search algorithm based on time window extending and time window shrinking strategies have been proposed. Key sequences from nonlinear correlation analysis are used in the next time series prediction model. Afterward, a novel dual-branch architecture that has synthesized the Structured Global Convolution (SGC) and iTransformer branches has been proposed which is called DBSGCformer. This framework enhances the ability to capture long-term dependencies through the combined effects of efficient convolution parameter optimization and variable-oriented multivariate modeling. We perform comprehensive experiments to investigate DBSGCformer’s potential in tackling complex multivariate time series forecasting challenges. Experiments conducted on two PV power datasets and five additional real-world datasets demonstrate that DBSGCformer significantly improves the accuracy of PV power forecasting and exhibits strong generalizability.
format Article
id doaj-art-c418a94aead545b7b5eaff8ece3ddaba
institution OA Journals
issn 0142-0615
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series International Journal of Electrical Power & Energy Systems
spelling doaj-art-c418a94aead545b7b5eaff8ece3ddaba2025-08-20T02:36:12ZengElsevierInternational Journal of Electrical Power & Energy Systems0142-06152025-08-0116911066010.1016/j.ijepes.2025.110660A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning frameworkMengji Yang0Haiqing Zhang1Xi Yu2Aicha Sekhari Seklouli3Abdelaziz Bouras4Yacine Ouzrout5Univ Lyon, Univ Lyon 2, INSA Lyon, Université Claude Bernard Lyon 1, DISP-UR4570, Bron, France; Stirling College, Chengdu University, Chengdu 610106, ChinaSchool of Software Engineering, Chengdu University of Information Technology, Chengdu 610225, China; Correspondence to: 24 Xuefu Road, 610225, Chengdu, China.Stirling College, Chengdu University, Chengdu 610106, ChinaUniv Lyon, Univ Lyon 2, INSA Lyon, Université Claude Bernard Lyon 1, DISP-UR4570, Bron, FranceDepartment of Computer Science, Qatar University, Doha, QatarUniv Lyon, Univ Lyon 2, INSA Lyon, Université Claude Bernard Lyon 1, DISP-UR4570, Bron, FranceKey factors influencing photovoltaic (PV) power generation predictions encompass solar radiation, aerosols, sunshine duration, temperature, humidity, wind direction, wind speed, cloud cover, and so on. The various influencing factors exhibit nonlinear correlation correlations, causing high volatility and discreteness in PV power time series. Firstly, to reduce the redundancy of the input for the prediction model and the computational time complexity, while enhancing the robustness and stability of the prediction model, nonlinear correlation search algorithm based on time window extending and time window shrinking strategies have been proposed. Key sequences from nonlinear correlation analysis are used in the next time series prediction model. Afterward, a novel dual-branch architecture that has synthesized the Structured Global Convolution (SGC) and iTransformer branches has been proposed which is called DBSGCformer. This framework enhances the ability to capture long-term dependencies through the combined effects of efficient convolution parameter optimization and variable-oriented multivariate modeling. We perform comprehensive experiments to investigate DBSGCformer’s potential in tackling complex multivariate time series forecasting challenges. Experiments conducted on two PV power datasets and five additional real-world datasets demonstrate that DBSGCformer significantly improves the accuracy of PV power forecasting and exhibits strong generalizability.http://www.sciencedirect.com/science/article/pii/S014206152500211XPhotovoltaic power predictionNonlinear correlation search algorithmStructured global convolutionHybrid prediction modeliTransformer
spellingShingle Mengji Yang
Haiqing Zhang
Xi Yu
Aicha Sekhari Seklouli
Abdelaziz Bouras
Yacine Ouzrout
A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
International Journal of Electrical Power & Energy Systems
Photovoltaic power prediction
Nonlinear correlation search algorithm
Structured global convolution
Hybrid prediction model
iTransformer
title A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
title_full A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
title_fullStr A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
title_full_unstemmed A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
title_short A composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
title_sort composite photovoltaic power prediction optimization model based on nonlinear meteorological factors analysis and hybrid deep learning framework
topic Photovoltaic power prediction
Nonlinear correlation search algorithm
Structured global convolution
Hybrid prediction model
iTransformer
url http://www.sciencedirect.com/science/article/pii/S014206152500211X
work_keys_str_mv AT mengjiyang acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT haiqingzhang acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT xiyu acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT aichasekhariseklouli acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT abdelazizbouras acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT yacineouzrout acompositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT mengjiyang compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT haiqingzhang compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT xiyu compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT aichasekhariseklouli compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT abdelazizbouras compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework
AT yacineouzrout compositephotovoltaicpowerpredictionoptimizationmodelbasedonnonlinearmeteorologicalfactorsanalysisandhybriddeeplearningframework