基于VMD分频与GRU-Informer协同机制的多风场风速预测

梁涛; 陈蕊; 米大斌; 井延伟; 郭艳旬; 张二辉

doi:10.19912/j.0254-0096.tynxb.2025-0712

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于VMD分频与GRU-Informer协同机制的多风场风速预测

更新时间：2026-10-15

- 基于VMD分频与GRU-Informer协同机制的多风场风速预测
- 基于VMD分频与GRU-Informer协同机制的多风场风速预测
- 太阳能学报 2026年47卷第9期页码：755-770
- 作者机构：
- 作者简介：
- 基金信息：
  
  河北省自然科学基金（E2024202051）;中央引导地方科技发展资金项目（264Z4304G）
- DOI：10.19912/j.0254-0096.tynxb.2025-0712
  中图分类号： TM614
- 网络首发：2026-10-10，
  
  纸质出版：2026
- 稿件说明：
移动端阅览
梁涛, 陈蕊, 米大斌, 等. 基于VMD分频与GRU-Informer协同机制的多风场风速预测[J]. 太阳能学报, 2026,47(9):755-770.

梁涛, 陈蕊, 米大斌, et al. 基于VMD分频与GRU-Informer协同机制的多风场风速预测[J]. 2026, 47(9): 755-770.
梁涛, 陈蕊, 米大斌, 等. 基于VMD分频与GRU-Informer协同机制的多风场风速预测[J]. 太阳能学报, 2026,47(9):755-770. DOI： 10.19912/j.0254-0096.tynxb.2025-0712.

梁涛, 陈蕊, 米大斌, et al. 基于VMD分频与GRU-Informer协同机制的多风场风速预测[J]. 2026, 47(9): 755-770. DOI： 10.19912/j.0254-0096.tynxb.2025-0712.

摘要

针对风电集控中心多风场风速序列非平稳性强及多尺度特征差异显著的问题

提出一种基于变分模态分解（VMD）分频与门控循环单元（GRU）-长序列时间序列预测模型（Informer）协同机制的混合风速预测模型。首先

利用VMD技术对典型风电场的风速及相关气象变量进行多模态分解

并将分解后的子序列重构为高频与低频两类信号。其次

构建分频协同预测架构

针对低频信号趋势性强、长期依赖显著的特点

采用GRU神经网络进行快速且精准的预测;针对高频信号波动剧烈、包含复杂细节的特点

利用Informer模型在长序列处理上的优势进行捕捉。随后

将两部分预测结果重组

形成典型风场的基础预测值。进一步地

针对集控中心多风场集中管控的需求

采用迁移学习策略将典型风场的特征模型迁移至中心平台

并引入多目标浣熊优化算法（MOCOA）对多源模型的预测结果进行自适应加权融合

构建具有强泛化能力的通用预测模型

以实现对辖区内其他风电场的精准预测。实验结果表明

该混合模型在风电集控中心的应用中展示了较高的预测精度、较强的环境适应能力及较快的响应速度

证明了其在实际应用中的有效性和优越性。

Abstract

To address the strong nonstationarity of wind speed time series

multi-scale variability

and cross-farm heterogeneity in centralized wind power control-center scenarios

this study proposes a hybrid wind speed forecasting model based on VMD-based frequency-band decomposition and a GRU-Informer synergistic mechanism. First

variational mode decomposition （VMD） is applied to wind speed and relevant meteorological variables from representative wind farms

and the decomposed modal components are reconstructed into low-frequency and high-frequency signals. A frequenly-band cooperative forecasting architecture is then developed. The gated recurrent unit （GRU） network is used to model low-frequency components with clear trend characteristics and long-term dependencies

while the Informer model is employed to capture complex fluctuations and long-range dependencies in high-frequency components. The forecasts from the two channels are subsequently reconstructed to obtain the baseline forecast for representative wind farms. Furthermore

to support multi-wind-farm forecasting under centralized management

a transfer learning strategy is introduced to transfer the learned source-domain models to target wind farms. A multi-objective coati optimization algorithm （MOCOA） is further used to adaptively fuse the forecasts of multiple source-domain expert models

thereby constructing a generalized forecasting model with improved adaptability and generalization capability. Experimental results show that the proposed hybrid model achieves high forecasting accuracy

strong environmental adaptability

and fast response in centralized wind power control applications

demonstrating its effectiveness and practical value.

关键词

Keywords

references

Chen Y H, Lin H Y.Overview of the development of offshore wind power generation in China[J]. Sustainable Energy Technologies and Assessments, 2022, 53: 102766.

Teng X Y, Zhuang W W, Liu F P, et al.China’s energy efficiency improvement on account of the development of wind and solar power: applying a dynamic non-radial directional distance function[J]. International Journal of Hydrogen Energy, 2022, 47(84): 35914-35927.

Yakoub G, Mathew S, Leal J.Direct and indirect short-term aggregated turbine- and farm-level wind power forecasts integrating several NWP sources[J]. Heliyon, 2023, 9(11): e21479.

Sheoran S, Pasari S.Efficacy and application of the window-sliding ARIMA for daily and weekly wind speed forecasting[J]. Journal of Renewable and Sustainable Energy, 2022, 14(5): 053305.

Zhang W Q, Lin Z, Liu X L.Short-term offshore wind power forecasting-a hybrid model based on discrete wavelet transform(DWT), seasonal autoregressive integrated moving average(SARIMA), and deep-learning-based long short-term memory(LSTM)[J]. Renewable Energy, 2022, 185: 611-628.

Ai X Y, Li S J, Xu H X.Wind speed prediction model using ensemble empirical mode decomposition, least squares support vector machine and long short-term memory[J]. Frontiers in Energy Research, 2023, 10: 1043867.

Zhu X X, Xu Z X, Wang Y, et al.Research on wind speed behavior prediction method based on multi-feature and multi-scale integrated learning[J]. Energy, 2023, 263: 125593.

Wu B R, Wang L.Two-stage decomposition and temporal fusion transformers for interpretable wind speed forecasting[J]. Energy, 2024, 288: 129728.

Yang J Y, Peng L, Luo L F, et al.Wind power forecasting using hybrid ICEEMDAN-KPCA and IWOA-BiLSTM models[J]. International Journal of Electrical Power & Energy Systems, 2025, 173: 111445.

Wang Y G, Zhao K X, Hao Y, et al.Short-term wind power prediction using a novel model based on butterfly optimization algorithm-variational mode decomposition-long short-term memory[J]. Applied Energy, 2024, 366: 123313.

Bengio Y, Simard P, Frasconi P.Learning long-term dependencies with gradient descent is difficult[J]. IEEE Transactions on Neural Networks, 1994, 5(2): 157-166.

Pascanu R,Mikolov T,Bengio Y.On the difficulty of training recurrent neural networks[C]//Proceedings of the 30th International Conference on Machine Learning. Atlanta, Georgia, USA, 2013,28(3): 1310-1318.

Vaswani A, Shazeer N, Parmar N, et al.Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA, 2017: 6000-6010.

Zhou H Y,Zhang S H,Peng J Q,et al.Informer: beyond efficient transformer for long sequence time-series forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2021,35(12): 11106-11115.

Lin C Y, Li X D, Shi T F, et al.Forecasting of wind speed under wind-fire coupling scenarios by combining HS-VMD and AM-LSTM[J]. Ecological Informatics, 2023, 77: 102270.

Wang X X, Shen X P, Ai X Y, et al.Short-term wind speed forecasting based on a hybrid model of ICEEMDAN, MFE, LSTM and informer[J]. PLoS One, 2023, 18(9): e0289161.

赵陆阳, 刘长良, 刘卫亮, 等. 基于CBAM-DSC-UNet模型的时空风速预测算法[J]. 太阳能学报, 2024, 45(10): 497-505.

孙亦皓, 刘浩, 胡天宇, 等. 基于时空关联特征与GCN-FEDformer的风速短期预测方法[J]. 中国电机工程学报, 2024, 44(21): 8496-8507.

Wang Y F, Yang Z H, Ma J H, et al.A wind speed forecasting framework for multiple turbines based on adaptive gate mechanism enhanced multi-graph attention networks[J]. Applied Energy, 2024, 372: 123777.

Al-Hajj R, Assi A, Neji B, et al.Transfer learning for renewable energy systems: a survey[J]. Sustainability, 2023, 15(11): 9131.

Li D, Hu Y, Yang B H, et al.A novel transfer learning strategy for wind power prediction based on TimesNet-GRU architecture[J]. Journal of Renewable and Sustainable Energy, 2024, 16(3): 033303.

邵宜祥, 刘剑, 胡丽萍, 等. 一种改进组合神经网络的超短期风速预测方法研究[J]. 发电技术, 2024, 45(2): 323-330.

Wang Y, Xu H H, Zou R M, et al.Dynamic non-constraint ensemble model for probabilistic wind power and wind speed forecasting[J]. Renewable and Sustainable Energy Reviews, 2024, 204: 114781.

Yan Y, Wang X R, Ren F, et al.Wind speed prediction using a hybrid model of EEMD and LSTM considering seasonal features[J]. Energy Reports, 2022, 8: 8965-8980.

刘洋, 王遂缠, 刘昊, 等. 基于WOA-VMD-CNN-Transformer的超短期风速组合预测[J]. 计算机系统应用, 2025, 34(12): 260-269.

Ayyavu S, Sayeed M S, Abdul Razak S F. A multi-step day-ahead wind power forecasting based on VMD-LSTM-EFG-ABC technique[J]. Energy Informatics, 2025, 8(1): 111.

Zhang H P, Wang J Z, Qian Y S, et al.Point and interval wind speed forecasting of multivariate time series based on dual-layer LSTM[J]. Energy, 2024, 294: 130875.

陈蕻峰, 王贺, 李岩, 等. 组合两步分解和ARIMA-LSTM的短期风速预测研究[J]. 太阳能学报, 2024, 45(2): 164-171.

Tang Q Z, Jiang Y, Xin J Z, et al.A novel method for the recovery of continuous missing data using multivariate variational mode decomposition and fully convolutional networks[J]. Measurement, 2023, 220: 113366.

Zhou S H, Gao C Y, Duan Z X, et al.A robust error correction method for numerical weather prediction wind speed based on Bayesian optimization, variational mode decomposition, principal component analysis, and random forest: VMD-PCA-RF (version 1.0.0)[J]. Geoscientific Model Development, 2023, 16(21): 6247-6266.

孔宪正, 黄国勇, 邓为权, 等. 基于SSA-VMD预处理的TCN-Informer短期风速多步预测混合模型[J]. 太阳能学报, 2026, 47(2): 527-538.

樊雅洁, 王聪, 张宏立, 等. 组合聚类和深度学习模型的风电场群风速预测[J]. 哈尔滨工业大学学报, 2024, 56(12): 71-80.

甘建红, 刘小锋, 白爱娟, 等. 基于时空特征融合的风速预测模型研究[J]. 微电子学与计算机, 2025, 42(7): 11-20.

Geng D H, Zhang Y K, Zhang Y L, et al.A hybrid model based on CapSA-VMD-ResNet-GRU-attention mechanism for ultra-short-term and short-term wind speed prediction[J]. Renewable Energy, 2025, 240: 122191.

伍耘, 葛佳敏, 王文烨, 等. 贝叶斯优化超参数的空时融合压缩残差网络在风速区间预测中的研究[J]. 电力系统保护与控制, 2025, 53(1): 13-23.

Xie A Q, Yang H, Chen J, et al.A short-term wind speed forecasting model based on a multi-variable long short-term memory network[J]. Atmosphere, 2021, 12(5): 651.

Li M D, Lei Z W, Ma Z W.Wind farm clustering method based on ASSCW-FCM algorithm[J]. Journal of Electrical Engineering & Technology, 2023, 18(1): 181-192.

Peng D G, Liu Y, Wang D H, et al.Short-term PV-Wind forecasting of large-scale regional site clusters based on FCM clustering and hybrid Inception-ResNet embedded with Informer[J]. Energy Conversion and Management, 2024, 320: 118992.

Dragomiretskiy K, Zosso D.Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.

Zhao Z N, Yun S N, Jia L Y, et al.Hybrid VMD-CNN-GRU-based model for short-term forecasting of wind power considering spatio-temporal features[J]. Engineering Applications of Artificial Intelligence, 2023, 121: 105982.

Hao Y, Wang X D, Wang J Z, et al.A new perspective of wind speed forecasting: multi-objective and model selection-based ensemble interval-valued wind speed forecasting system[J]. Energy Conversion and Management, 2024, 299: 117868.

Dehghani M, Montazeri Z, Trojovská E, et al.Coati optimization algorithm: a new bio-inspired metaheuristic algorithm for solving optimization problems[J]. Knowledge-Based Systems, 2023, 259: 110011.

Liang T, Chai C J, Sun H X, et al.Wind speed prediction based on multi-variable Capsnet-BILSTM-MOHHO for WPCCC[J]. Energy, 2022, 250: 123761.

Gong Z P, Wan A P, Ji Y S, et al.Improving short-term offshore wind speed forecast accuracy using a VMD-PE-FCGRU hybrid model[J]. Energy, 2024, 295: 131016.

Bommidi B S, Teeparthi K, Kosana V.Hybrid wind speed forecasting using ICEEMDAN and transformer model with novel loss function[J]. Energy, 2023, 265: 126383.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于WRF模式的中长期风速预报及订正研究

风资源评估测风数据插补方法的时空特性研究

优化变分模态分解下NRBO-LSTM-Attention修正预测风速的风电功率短期预测

基于时频插值的非平稳风速随机波模拟

基于机器学习方法的风速偏差订正研究