Probabilistic Power Forecasting Method for Distributed Photovoltaic Clusters Based on Similar Cloud Condition Matching

GONG Mingkai; LI Kangping; LI Zhenghui; HUANG Chunyi

doi:10.7500/AEPS20250509007

您当前的位置：

首页 >

文章列表页 >

Probabilistic Power Forecasting Method for Distributed Photovoltaic Clusters Based on Similar Cloud Condition Matching

Basic Research | 更新时间：2026-02-28

- Probabilistic Power Forecasting Method for Distributed Photovoltaic Clusters Based on Similar Cloud Condition Matching
- Automation of Electric Power Systems Vol. 50, Issue 2, Pages: 93-102(2026)
- 作者机构：
  
  1.上海交通大学智慧能源创新学院，上海市 200240
  2.上海非碳基能源转换与利用研究院，上海市 200240
  3.上海交通大学电气工程学院，上海市 200240
- 作者简介：
- 基金信息：
  
  National Key R&D Program of China(2024YFE0106900)
- DOI：10.7500/AEPS20250509007
  CLC：
- Received：09 May 2025，
  
  Revised：2025-07-31，
  
  Online First：23 September 2025，
  
  Published：25 January 2026
- 稿件说明：
移动端阅览
龚明凯,李康平,李正辉等.基于相似云况匹配的分布式光伏集群功率概率预测方法[J].电力系统自动化,2026,50(2):93-102.

GONG Mingkai,LI Kangping,LI Zhenghui,et al.Probabilistic Power Forecasting Method for Distributed Photovoltaic Clusters Based on Similar Cloud Condition Matching[J].Automation of Electric Power Systems,2026,50(2):93-102.
龚明凯,李康平,李正辉等.基于相似云况匹配的分布式光伏集群功率概率预测方法[J].电力系统自动化,2026,50(2):93-102. DOI： 10.7500/AEPS20250509007.

GONG Mingkai,LI Kangping,LI Zhenghui,et al.Probabilistic Power Forecasting Method for Distributed Photovoltaic Clusters Based on Similar Cloud Condition Matching[J].Automation of Electric Power Systems,2026,50(2):93-102. DOI： 10.7500/AEPS20250509007.

摘要

分布式光伏系统缺乏现场气象实测数据，而大范围、高精度的数值天气预报数据获取成本高昂，这使得现有针对集中式光伏的功率预测方法难以适用。卫星云图能够低成本地提供大范围、高分辨率的云层观测信息，有望成为解决该问题的重要气象信息来源。为此，提出了一种基于相似云况匹配的分布式光伏集群功率概率预测方法。首先，构建了光流增强的三维卷积神经网络来预测未来的卫星云图，该模型通过引入物理一致性损失引导模型关注云团的动态演变过程，从而提升预测结果的物理合理性；其次，提出了一种相似云况匹配方法，通过从历史中寻找与待预测时段最相似的云况并提取对应的光伏功率作为后续功率预测的输入特征；最后，将以上述相似历史功率与邻近时刻的实测光伏功率等常规信息联合作为样条分位数回归模型的输入，实现光伏功率的准确概率预测。在包含600多个分布式光伏站点的大规模真实数据集上进行算例分析，结果表明所提出的方法在连续排名概率评分指标上相比经典的核密度估计方法提高显著。

Abstract

Distributed photovoltaic (PV) systems lack on-site meteorological measurements

while the cost of acquiring large-scale and high-resolution numerical weather prediction (NWP) data is very high

which makes it difficult for existing power forecasting methods for centralized photovoltaics to be applicable. Satellite cloud images

offering wide-area

high-resolution cloud observations at a low cost

which is expected to become an important source of meteorological information to solve this problem. Therefore

this paper proposes a probabilistic power forecasting method for distributed PV clusters based on similar cloud condition matching. First

an optical-flow-enhanced 3-dimensional convolutional neural network (3D-CNN) is constructed to forecast future satellite cloud images. By incorporating a physical consistency loss

the model is guided to focus on the dynamic evolution of cloud structures

thereby improving the physical plausibility of the forecasting results. Secondly

a cloud condition matching method is introduced

by searching for the cloud conditions that are most similar to the forecasting period in history

the corresponding photovoltaic power is extracted as input features for subsequent power forecasting. Finally

the common information such as similar historical power and measured PV power at adjacent times will be combined as inputs for the spline quantile regression model to achieve accurate probabilistic forecasting of PV power. Case studies are conducted on a large-scale real dataset containing over 600 distributed PV sites

and the results show that the proposed method shows significant improvement compared to classical kernel density estimation methods in terms of the continuous ranked probability score index.

关键词

Keywords

references

国家能源局 . 国家能源局发布2024年全国电力工业统计数据［EB/OL］.（ 2025-01-21 ）［ 2025-03-17 ］. https：//www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html .

National Energy Administration . The National Energy Administration has released statistical data on the national power industry for 2024 ［EB/OL］. （ 2025-01-21 ）［ 2025-03-17 ］. https：//www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html .

LI K P ， GONG M K ， LI Z H ， et al . Classified identification and estimation of behind-the-meter distributed photovoltaic panels using high-resolution aerial imagery ［J］. IEEE Transactions on Industry Applications ， 2025 ， 61 （ 1 ）： 1604 - 1612 .

万灿，宋永华 . 新能源电力系统概率预测理论与方法及其应用［J］. 电力系统自动化， 2021 ， 45 （ 1 ）： 2 - 16 .

WAN Can ， SONG Yonghua . Theories， methodologies and applications of probabilistic forecasting for power systems with renewable energy sources ［J］. Automation of Electric Power Systems ， 2021 ， 45 （ 1 ）： 2 - 16 .

万灿，崔文康，宋永华 . 新能源电力系统概率预测：基本概念与数学原理［J］. 中国电机工程学报， 2021 ， 41 （ 19 ）： 6493 - 6509 .

WAN Can ， CUI Wenkang ， SONG Yonghua . Probabilistic forecasting for power systems with renewable energy sources： basic concepts and mathematical principles ［J］. Proceedings of the CSEE ， 2021 ， 41 （ 19 ）： 6493 - 6509 .

WAN C ， LIN J ， WANG J H ， et al . Direct quantile regression for nonparametric probabilistic forecasting of wind power generation ［J］. IEEE Transactions on Power Systems ， 2017 ， 32 （ 4 ）： 2767 - 2778 .

WAN C ， XU Z ， PINSON P . Direct interval forecasting of wind power ［J］. IEEE Transactions on Power Systems ， 2013 ， 28 （ 4 ）： 4877 - 4878 .

YU G Z ， LU L ， TANG B ， et al . Ultra-short-term wind power subsection forecasting method based on extreme weather ［J］. IEEE Transactions on Power Systems ， 2023 ， 38 （ 6 ）： 5045 - 5056 .

WAN C ， LIN J ， SONG Y H ， et al . Probabilistic forecasting of photovoltaic generation： an efficient statistical approach ［J］. IEEE Transactions on Power Systems ， 2017 ， 32 （ 3 ）： 2471 - 2472 .

林帆，张耀，东琦，等 . 基于分位数插值和深度自回归网络的光伏出力概率预测［J］. 电力系统自动化， 2023 ， 47 （ 9 ）： 79 - 87 .

LIN Fan ， ZHANG Yao ， DONG Qi ， et al . Probability prediction of photovoltaic output based on quantile interpolation and deep autoregressive network ［J］. Automation of Electric Power Systems ， 2023 ， 47 （ 9 ）： 79 - 87 .

张昆明，马龙义，章天晗，等 . 基于iDGA-LSTM的分布式光伏出力时空协同概率预测方法［J］. 电力系统自动化， 2025 ， 49 （ 1 ）： 128 - 138 .

ZHANG Kunming ， MA Longyi ， ZHANG Tianhan ， et al . Spatio-temporal cooperative probability forecasting method for distributed photovoltaic output based on iDGA-LSTM ［J］. Automation of Electric Power Systems ， 2025 ， 49 （ 1 ）： 128 - 138 .

赵耀，高少炜，李东东，等 . 基于天气相似聚类与QRNN的短期光伏功率区间概率预测［J］. 电力系统自动化， 2023 ， 47 （ 23 ）： 152 - 161 .

ZHAO Yao ， GAO Shaowei ， LI Dongdong ， et al . Short-term interval probability prediction of photovoltaic power based on weather similarity clustering and quantile regression neural network ［J］. Automation of Electric Power Systems ， 2023 ， 47 （ 23 ）： 152 - 161 .

董雷，周文萍，张沛，等 . 基于动态贝叶斯网络的光伏发电短期概率预测［J］. 中国电机工程学报， 2013 ， 33 （增刊1 ）： 38 - 45 .

DONG Lei ， ZHOU Wenping ， ZHANG Pei ， et al . Short-term photovoltaic output forecast based on dynamic Bayesian network theory ［J］. Proceedings of the CSEE ， 2013 ， 33 （ S1 ）： 38 - 45 .

路志英，任一墨，葛路琨 . 基于样条估计分位数回归的光伏功率回归模型［J］. 湖南大学学报（自然科学版）， 2017 ， 44 （ 10 ）： 91 - 98 .

LU Zhiying ， REN Yimo ， GE Lukun . Photovoltaic power regression model based on spline estimation and quantile regression ［J］. Journal of Hunan University （Natural Sciences）， 2017 ， 44 （ 10 ）： 91 - 98 .

WANG K ， ZHANG Y ， LIN F ， et al . Nonparametric probabilistic forecasting for wind power generation using quadratic spline quantile function and autoregressive recurrent neural network ［J］. IEEE Transactions on Sustainable Energy ， 2022 ， 13 （ 4 ）： 1930 - 1943 .

GASTHAUS J ， BENIDIS K ， WANG Y ， et al . Probabilistic forecasting with spline quantile function RNNs ［C］// Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics ， April 16-18 ， 2019 ， Okinawa， Japan .

BESSA R J ， MIRANDA V ， BOTTERUD A ， et al . Time adaptive conditional kernel density estimation for wind power forecasting ［J］. IEEE Transactions on Sustainable Energy ， 2012 ， 3 （ 4 ）： 660 - 669 .

SI Z Y ， YU Y X ， YANG M ， et al . Hybrid solar forecasting method using satellite visible images and modified convolutional neural networks ［J］. IEEE Transactions on Industry Applications ， 2021 ， 57 （ 1 ）： 5 - 16 .

余光正，陆柳，汤波，等 . 基于云图特征提取的改进混合神经网络超短期光伏功率预测方法［J］. 中国电机工程学报， 2021 ， 41 （ 20 ）： 6989 - 7003 .

YU Guangzheng ， LU Liu ， TANG Bo ， et al . An improved hybrid neural network ultra-short-term photovoltaic power forecasting method based on cloud image feature extraction ［J］. Proceedings of the CSEE ， 2021 ， 41 （ 20 ）： 6989 - 7003 .

陈殿昊，臧海祥，蒋雨楠，等 . 基于多层次云图特征与宽度学习的超短期光伏功率预测［J］. 电力系统自动化， 2024 ， 48 （ 22 ）： 131 - 139 .

CHEN Dianhao ， ZANG Haixiang ， JIANG Yunan ， et al . Ultra-short-term photovoltaic power forecasting based on multi-level sky image features and broad learning ［J］. Automation of Electric Power Systems ， 2024 ， 48 （ 22 ）： 131 - 139 .

SAKAINO H . Spatio-temporal image pattern prediction method based on a physical model with time-varying optical flow ［J］. IEEE Transactions on Geoscience and Remote Sensing ， 2013 ， 51 （ 5 ）： 3023 - 3036 .

TEED Z ， DENG J . RAFT： recurrent all-pairs field transforms for optical flow ［M］. Switzerland， Cham ： Springer International Publishing ， 2020 ： 402 - 419 .

国家卫星气象中心 . FY-4A AGRI LI_4KM 数据产品［DB/OL］.［ 2025-04-30 ］. https：//satellite.nsmc.org.cn/DataPortal/cn/data/structure.html https://satellite.nsmc.org.cn/DataPortal/cn/data/structure.html .

National Satellite Meteorological Center . FY-4A AGRI LI_4KM data product ［DB/OL］. ［ 2025-04-30 ］. https：//satellite.nsmc.org.cn/DataPortal/cn/data/structure.html https://satellite.nsmc.org.cn/DataPortal/cn/data/structure.html .

国家卫星气象中心 . FY-4A 地表温度产品［DB/OL］.［ 2025-04-30 ］. https：//satellite.nsmc.org.cn/DataPortal/cn/data/structure.html https://satellite.nsmc.org.cn/DataPortal/cn/data/structure.html .

National Satellite Meteorological Center . FY-4A surface temperature product ［DB/OL］. ［ 2025-04-30 ］. https：//satellite.nsmc.org.cn/DataPortal/cn/data/structure.html https://satellite.nsmc.org.cn/DataPortal/cn/data/structure.html .

BRÖCKER J ， SMITH L A . Scoring probabilistic forecasts： the importance of being proper ［J］. Weather and Forecasting ， 2007 ， 22 （ 2 ）： 382 - 388 .

WU X Y ， ZHEN Z ， ZHANG J M ， et al . Multidimensional feature extraction based minutely solar irradiance forecasting method using all-sky images ［J］. IEEE Transactions on Industry Applications ， 2024 ， 60 （ 3 ）： 4494 - 4504 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Enhancement Method for Photovoltaic Accommodation Capacity of Distribution Networks Based on Self-attention Diffusion Reinforcement Learning

Distributed Unbalanced Current Compensation Control in Distribution Network with Participation of Photovoltaic

Two-stage Centralized-Distributed Voltage Control for Distribution Networks Based on Deep Reinforcement Learning

Joint Estimation of Lithium-ion Battery State Based on Light Gradient Boosting Machine

Related Author

YANG Wenwei

PENG Xiangang

CAI Weicong

OUYANG Sheng

WANG Xinghua

ZHAO Zhuoli

FENG Qifan

XIA Yanghong

Related Institution

College of Automation, Guangdong University of Technology

College of Electrical Engineering, Zhejiang University

School of Information and Electrical Engineering, Hangzhou City University

College of Electrical and Information Engineering, Hunan University

China Electric Power Research Institute Co., Ltd.

⁰