Author Login Chief Editor Login Reviewer Login Editor Login Remote Office

Computer Engineering ›› 2026, Vol. 52 ›› Issue (10): 50-72. doi: 10.19678/j.issn.1000-3428.0253044

• Frontier Perspectives and Reviews • Previous Articles     Next Articles

A Review of Deep Learning Applications in Short-term Photovoltaic Power Forecasting

DING Tongguang1, DU Shengdong2,*(), ZHAO Han3, QIAN Wei1, GUO Chushan2, LIU Fan3   

  1. 1. Datang Hami New Energy Co., Ltd., Hami 839099, Xinjiang, China
    2. Northwest Electric Power Test Institute of China Datang Group Science and Technology Research Institute Co., Ltd., Xi'an 710018, Shaanxi, China
    3. College of Computer Science and Software Engineering, Hohai University, Nanjing 211100, Jiangsu, China
  • Received:2025-09-19 Revised:2025-12-15 Online:2026-10-15 Published:2026-10-26
  • Contact: DU Shengdong

面向短期光伏功率预测的深度学习应用综述

丁同光1, 杜胜东2,*(), 赵晗3, 钱为1, 郭楚珊2, 刘凡3   

  1. 1. 大唐哈密新能源有限公司, 新疆 哈密 839099
    2. 中国大唐集团科学技术研究总院有限公司西北电力试验研究院, 陕西 西安 710018
    3. 河海大学计算机与软件学院, 江苏 南京 211100
  • 通讯作者: 杜胜东
  • 作者简介:

    丁同光, 男, 高级工程师, 主研方向为机械设计制造及其自动化

    杜胜东(通信作者), 工程师

    赵晗, 讲师、博士

    钱为, 高级工程师

    郭楚珊, 工程师

    刘凡, 教授、博士

  • 基金资助:
    国家自然科学基金面上项目(62372155)

Abstract:

Short-term photovoltaic power forecasting is the foundation for optimizing power dispatch in power systems, and its accuracy directly affects the overall system efficiency. However, photovoltaic power, which is affected by multiple meteorological factors, exhibits short-term volatility and randomness, posing severe challenges for high-precision forecasting. In recent years, Deep Learning (DL) has shown an excellent ability to excavate intrinsically correlated features, offering a promising technical path for precise forecasting. The application of DL to short-term photovoltaic power forecasting was studied systematically. First, a typical application paradigm for mainstream deep learning models is presented. Subsequently, the application status and performance of deep neural networks, such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformers, and Graph Neural Networks (GNNs), in the task are discussed considering the ideal scenario of sufficient available data with static and no incremental changes. Furthermore, the strategies and progress of deep neural network based on technologies such as deep data augmentation, Transfer Learning (TL), Federated Learning (FL), and Online Learning (OL) are analyzed to address real-world challenges such as data scarcity and limited access. Finally, the challenges regarding robustness, generalization, and adaptation faced by existing research are discussed along with future research routes from the perspectives of forecasting architecture and optimization strategies.

Key words: Deep Learning (DL), short-term photovoltaic power forecasting, deep neural network, deep data augmentation, Transfer Learning (TL), Federated Learning (FL)

摘要:

短期光伏功率预测是电力系统实现优化调度的核心基础, 其精度直接影响系统整体效能。然而, 光伏发电功率受多重气象因素影响, 所呈现出的短期波动性和随机性给高精度预测的实现带来严峻挑战。近年来, 深度学习(DL)技术凭借其挖掘数据内在关联特征的卓越能力, 为突破性提升短期光伏功率预测精度开辟了全新路径。全面梳理现有DL技术在短期光伏功率预测任务中的应用进展。首先, 阐述DL在该预测任务中的典型应用范式; 然后, 介绍在可用数据充足、静态且无增量的理想场景下, 卷积神经网络(CNN)、循环神经网络(RNN)、转换器(Transformer)及图神经网络(GNN)等深度神经网络在该任务中的应用现状; 继而, 剖析基于深度数据增强、迁移学习(TL)、联邦学习(FL)、在线学习(OL)等技术的深度神经网络在应对数据匮乏、访问受限等现实数据挑战时, 在该任务中的应用思路与进展; 最后, 探讨当前基于DL的短期光伏功率预测研究在鲁棒性、泛化性、适应性等方面面临的挑战, 并从模型架构、优化策略等角度对未来技术路线进行前瞻性展望。

关键词: 深度学习, 短期光伏功率预测, 深度神经网络, 深度数据增强, 迁移学习, 联邦学习