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

作　　者：赵雪锋 张宇宁 詹巍 李明烜 杨岩 杨磊[3] 杨锡运[3] ZHAO Xuefeng;ZHANG Yuning;ZHAN Wei;LI Mingxuan;YANG Yan;YANG Lei;YANG Xiyun(State Power Investment Group Sichuan Electric Power Co.,Ltd.,Chengdu 610213,Sichuan,China;State Power Investment Group Southwest Energy Research Institute Co.,Ltd.,Chengdu 610213,Sichuan,China;School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]国家电投集团四川电力有限公司,四川成都610213 [2]国家电投集团西南能源研究院有限公司,四川成都610213 [3]华北电力大学控制与计算机工程学院,北京102206

出　　处：《水力发电》2025年第3期97-104,124,共9页Water Power

基　　金：四川省科技计划重点研发项目(2023YFG0108);国家电投集团四川电力有限公司科技项目(XNNY-WW-KJ-2021-16)。

摘　　要：为解决光伏功率气象源数据质量不佳和来源单一所造成的预测精度不足问题,提出了一种基于TCN-Informer的多源融合虚拟最优气象预报源的光伏功率组合预测模型。首先,选取待预测电站临近气象站点,提取功率预报相关气象数据,筛选最佳气象因素;随后,通过数据融合构建光伏电站虚拟最优预报气象源,提高预测模型输入数据质量;最后,提出了一种基于TCN-Informer的功率预测模型,利用TCN的局部特征提取能力,对Informer提取的长期依赖关系和全局特征进行细化,实现光伏功率的精准预测。光伏电站实际算例验证表明,基于TCN-Informer的多源融合虚拟最优气象预报源光伏功率预测模型在4种不同天气类别下均具有较高的预测精度和较好的鲁棒性。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 

分 类 号：TM615[电气工程—电力系统及自动化]