山东电力技术2025，Vol.52Issue(12)：17-26,10.DOI:10.20097/j.cnki.issn1007-9904.240618
基于地基云图与卫星云图相结合的光伏功率预测方法
Photovoltaic Power Prediction Method Based on Combination of Ground-based and Satellite Cloud Maps
摘要
Abstract
As one of the richest sources of renewable energy,solar energy holds significant development potential,especially in the current context of China's energy structure adjustment.Its development and utilization is particularly important.Photovoltaic(PV)power generation is one of the most widely employed methods for harnessing solar energy.However,the efficiency and stability of PV power generation are significantly affected by natural factors,such as irradiance levels and variations in cloud cover,which can lead to diurnal fluctuations in PV power.In order to improve the accuracy of PV power prediction,this paper proposes a novel PV power prediction method that combines satellite cloud maps and ground-based cloud maps.The method utilizes a convolutional neural network to extract features from both satellite and ground-based cloud maps,enabling ultra-short-term predictions of PV power output.By accurately capturing the dynamic changes in cloud cover,the method can effectively alleviate the pressure on the power grid caused by PV power fluctuations due to natural factors.Furthermore,it provides robust technical support for the widespread adoption of PV power generation technology and the intelligent management of the power grid.关键词
卫星云图/地基云图/卷积神经网络/光伏功率Key words
satellite cloud map/ground-based maps/convolutional neural network/photovoltaic power分类
信息技术与安全科学引用本文复制引用
MA Wenwen,HU Siyu,YANG Fan,LI Dengxuan,BU Qiangsheng..基于地基云图与卫星云图相结合的光伏功率预测方法[J].山东电力技术,2025,52(12):17-26,10.基金项目
国家电网公司总部管理科技项目"基于分布式光伏多层级预测的配电网运行风险预控关键技术研究及应用"(5400-202355555A-3-2-ZN).Science and Technology Project of State Grid Corporation of China"Research and Application of Key Technologies for Operation Risk Pre-control of Distribution Network Based on Distributed Photovoltaic Multilayer Prediction"(5400-202355555A-3-2-ZN). （5400-202355555A-3-2-ZN）
