水力发电2025，Vol.51Issue(3)：97-104,124,9.
基于TCN-Informer的多源融合虚拟最优气象源光伏功率预测
Photovoltaic Power Prediction Based on TCN-Informer and Multi-source Fusion of Virtual Optimal Weather Sources
摘要
Abstract
In order to solve the problem of insufficient prediction accuracy caused by the poor quality of PV power weather source data and a single source,a PV power combination prediction model based on TCN-Informer and multi-source fusion of virtual optimal weather forecast source is proposed.Firstly,the meteorological stations near the power station to be predicted are selected to extract the power forecast-related meteorological data,and the best meteorological factors are screened.Subsequently,the virtual optimal forecast meteorological source of PV power station is constructed through data fusion to improve the quality of the input data for the prediction model.Finally,a power prediction model based on TCN-Informer is proposed,which takes advantage of TCN's ability of localized feature extraction to refine the Informer extracted long-term dependencies and global features,so as to achieve accurate prediction of PV power.The actual case validation of PV power stations shows that the PV power prediction model based on TCN-Informer and multi-source fusion virtual optimal weather forecast source has higher prediction accuracy and good robustness under four different weather categories.关键词
多源融合/虚拟最优/TCN/Informer/光伏功率预测Key words
multi-source fusion/virtual optimization/TCN/Informer/photovoltaic power prediction分类
信息技术与安全科学引用本文复制引用
赵雪锋,张宇宁,詹巍,李明烜,杨岩,杨磊,杨锡运..基于TCN-Informer的多源融合虚拟最优气象源光伏功率预测[J].水力发电,2025,51(3):97-104,124,9.基金项目
四川省科技计划重点研发项目(2023YFG0108) （2023YFG0108）
国家电投集团四川电力有限公司科技项目(XNNY-WW-KJ-2021-16) （XNNY-WW-KJ-2021-16）
