Photovoltaic Power Forecasting with Weather Conditioned Attention Mechanism

作     者：Xuetao Jiang Yuchun Gou Meiyu Jiang Lihui Luo Qingguo Zhou 

作者机构：School of Information Science and EngineeringLanzhou UniversityLanzhou 730000China Gansu Provincial Industry Technology Center of Intelligent Equipment&Big Data for Disaster PreventionNorthwest Institute of Eco-Environmentand ResourcesChinese Academy of SciencesLanzhou 730000China 

出 版 物：《Big Data Mining and Analytics》 (大数据挖掘与分析(英文))

年 卷 期：2025年第8卷第2期

页      面：326-345页

核心收录：

学科分类：08[工学] 0807[工学-动力工程及工程热物理] 

基　　金：supported by the National Natural Science Foundation of China(No.U22A20261) National Key R&D Program of China(No.2023YFB4503903) Gansu Province Science and Technology Major Project—Industrial Project(Nos.22ZD6GA048 and 23ZDGA006) Gansu Province Key Research and Development Plan—Industrial Project(No.22YF7GA004) Fundamental Research Funds for the Central Universities(No.lzujbky-2022-kb12) Open Project of Gansu Provincial Key Laboratory of Intelligent Transportation(No.GJJ-ZH-2024-002) Gansu Provincial Science and Technology Major Special Innovation Consortium Project(No.21ZD3GA002) Science and Technology Plan of Qinghai Province(No.2020-GX164) OPPO Research Fund,Supercomputing Center of Lanzhou University 

主　　题：Photovoltaic(PV)forecasting deep learning transformer Desert Knowledge Australia Solar Centre(DKASC) 

摘      要：Accurate Photovoltaic(PV)generation forecasts can reduce power redeploy from the grid,thus increasing the supplier’s profit in the day-ahead electricity ***,the PV process is affected differently by various factors under different weather conditions,resulting in significantly different energy output *** this context,this paper proposes a day-ahead PV power forecasting method with weather conditioned attention *** propose a Multi-Stream Attention Fusion Network(MSAFN)which utilizes an algorithm to derive the optimal decomposition algorithm for different weather *** proposed Conditional Decomposition(CD)algorithm searches for the decomposition algorithms and corresponding hyperparameters of the prediction model,aiming to achieve the optimal prediction *** MSAFN incorporates multiple attention modules to learn the energy output patterns under various weather ***,the attention modules adeptly learn patterns under diverse conditions,while simultaneously,the sharing of weights among the remaining components of the model effectively enhances prediction accuracy and facilitates a reduction in training *** compare the state-of-the-art decomposition algorithms(VMD,EEMD,MSTL,etc.)and prediction models(BPN,LSTM,XGBoost,transformer,etc.)commonly used in PV *** results show that the MSAFN model is more accurate than the models above,which has a noticeable improvement compared to other recent day-ahead PV predictions on Desert Knowledge Australia Solar Centre(DKASC)dataset.