发电技术 ›› 2026, Vol. 47 ›› Issue (1): 53-64.DOI: 10.12096/j.2096-4528.pgt.260105

• 新能源 • 上一篇    下一篇

考虑天气耦合相似日的短期光伏功率预测

逯静1, 杨源浩2, 汪中宏3, 王瑞2   

  1. 1.河南理工大学计算机科学与技术学院,河南省 焦作市 454000
    2.河南理工大学电气 工程与自动化学院,河南省 焦作市 454000
    3.华能日照电厂,山东省 日照市 276826
  • 收稿日期:2025-01-17 修回日期:2025-04-24 出版日期:2026-02-28 发布日期:2026-02-12
  • 作者简介:逯静(1980),女,硕士,副教授,研究方向为新能源功率预测、深度学习和物联网技术等,lujing@hpu.edu.cn
    杨源浩(2000),男,硕士研究生,研究方向为新能源功率预测,1924532335@qq.com
    汪中宏(1977),男,高级工程师,研究方向为火电机组集控运行、锅炉检修及新能源运维管理,wzhsywsc@163.com
    王瑞(1977),男,硕士,副教授,研究方向为电力系统分析和智能信息处理,wangrui@hpu.edu.cn
  • 基金资助:
    国家自然科学基金项目(62273133);河南省科技攻关项目(222102210120)

Short-Term PV Power Prediction Considering Weather-Coupled Similar Days

Jing LU1, Yuanhao YANG2, Zhonghong WANG3, Rui WANG2   

  1. 1.School of Computer Science and Technology, Henan Polytechnic University, Jiaozuo 454000, Henan Province, China
    2.School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo 454000, Henan Province, China
    3.Huaneng Rizhao Power Plant, Rizhao 276826, Shandong Province, China
  • Received:2025-01-17 Revised:2025-04-24 Published:2026-02-28 Online:2026-02-12
  • Supported by:
    National Natural Science Foundation of China(62273133);Henan Science and Technology Research Project(222102210120)

摘要:

目的 为充分利用历史信息,最大限度地优化模型效果,提高光伏功率预测精度,提出了一种考虑天气耦合相似日的短期光伏功率预测方法。 方法 首先,利用模糊C均值聚类将数据集划分为不同天气类型,根据待测日对每个天气类型的隶属度和特征选择计算关联度权重因子,结合灰色关联分析计算历史日的相似度并筛选具有天气耦合的相似日集,通过变分模态分解将相似日集分解为不同频率的模态分量,实现进一步去噪。其次,为充分发挥模型非线性拟合能力,运用红尾鹰算法(red-tailed hawk algorithm,RTHA)对双向长短时记忆(bidirectional long short-term memory,BiLSTM)网络模型进行超参数寻优,并构建RTHA-BiLSTM模型对各个模态分量进行预测。最后,以我国江苏某电厂的实际数据为例进行仿真实验,验证所提方法的有效性。 结果 在晴天、多云和雨天场景下,与无相似日方法相比,所提方法在单一模型和组合模型中均方根误差分别降低了9.1%、6.1%、2.9%和11.1%、6.5%、13.9%。 结论 所提方法可有效提升光伏功率预测精度,具有较好的鲁棒性和较强的预测能力,能较好地应对不同场景下的预测任务。

关键词: 光伏(PV)发电, 功率预测, 模糊C均值聚类, 灰色关联分析, 相似日, 红尾鹰算法

Abstract:

Objectives To make full use of historical information, maximize the optimization of model effect and improve the accuracy of photovoltaic (PV) power prediction, a short-term PV power prediction method considering weather-coupled similar days is proposed. Methods Firstly, fuzzy C-means clustering is used to divide the dataset into different weather types, and the correlation weight factors are calculated according to the affiliation and features selection of the day to be predicted for each weather type, and the similarity of the historical days is calculated by grey relation analysis, and similar days with weather coupling are selected. The similar days are decomposed into modal components with different frequencies by variational mode decomposition to realize further denoising. Secondly, in order to fully utilize the nonlinear fitting ability of the model, the red-tailed hawk algorithm (RTHA) is used to optimize the hyperparameters of the bidirectional long short-term memory (BiLSTM) network model, and the RTHA-BiLSTM model is constructed to predict the modal components. Finally, taking the actual data of a power plant in Jiangsu Province as an example, the simulation experiment is carried out to verify the effectiveness of the proposed method. Results In sunny, cloudy and rainy scenarios, compared with the method without similar days, the proposed method reduces the root mean square error by 9.1%, 6.1%, 2.9% and 11.1%, 6.5%, 13.9% in the single model and the combined model, respectively. Conclusions The proposed method can effectively improve the prediction accuracy of PV power, has good robustness and strong prediction ability, and can better cope with the prediction tasks in different scenarios.

Key words: photovoltaic (PV) power generation, power prediction, fuzzy C-means clustering, grey relation analysis, similar days, red-tailed hawk algorithm

中图分类号: