基于VMD-LSTM-MultiConv-SelfAttention的超短期风电功率预测研究

任浩琴; 连维畅; 亓凤梧; 王立敏; 赵树涵; 刘广臣

doi:10.19912/j.0254-0096.tynxb.2025-0170

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

基于VMD-LSTM-MultiConv-SelfAttention的超短期风电功率预测研究

更新时间：2026-07-22

- 基于VMD-LSTM-MultiConv-SelfAttention的超短期风电功率预测研究
- Vol. 47, Issue 6, Pages: 394-402(2026)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.19912/j.0254-0096.tynxb.2025-0170
  CLC： TM614
- Online First：08 July 2026，
  
  Published：2026
- 稿件说明：
移动端阅览
任浩琴, 连维畅, 亓凤梧, et al. 基于VMD-LSTM-MultiConv-SelfAttention的超短期风电功率预测研究[J]. 2026, 47(6): 394-402.
DOI：

任浩琴, 连维畅, 亓凤梧, et al. 基于VMD-LSTM-MultiConv-SelfAttention的超短期风电功率预测研究[J]. 2026, 47(6): 394-402. DOI： 10.19912/j.0254-0096.tynxb.2025-0170.

摘要

为提高风电功率预测精度同时确保电力系统的高效调度

提出一种多算法集成的超短期风电功率预测模型

通过变分模态分解（VMD）对功率序列进行噪声抑制与重构。在模型构建阶段

利用LSTM捕捉时序依赖

其后连接多层卷积神经网络提取局部特征并引入自注意力机制动态聚焦关键时间步

形成多模块协同的预测框架。为检验模型的实效

以山东省某风电场为核心开展消融、对比及季节迁移实验。结果显示：相较于基准模型

该模型的平均绝对误差MAE降低23.5%

均方根误差RMSE减少20%

可凸显各模块在捕捉复杂时序特征中的优势。进一步验证表明

模型在跨区域（中西部风电场）与跨能源（光伏电站）场景中均表现优异

凸显其强泛化能力。

Abstract

To enhance the accuracy of ultra-short-term wind power forecasting and support efficient power system dispatch

this study proposes an integrated multi-algorithm forecasting model. The variational mode decomposition （VMD） is firstly applied to suppress noise and reconstruct the original power sequence. In the modeling stage

a long short-term memory （LSTM） network is adopted to capture temporal dependencies

followed by a multi-layer convolutional neural network （CNN） to extract local features. A Self Attention mechanism is further incorporated to dynamically focus on critical time steps

resulting in a collaborative multi-module forecasting framework. To evaluate the performance of the proposed model

ablation studies

comparative experiments

and seasonal transfer tests were conducted using data from a wind farm in Shandong Province

China. The results show that

compared to baseline models

the proposed model reduces the mean absolute error （MAE） by 23.5% and the root mean square error （RMSE） by 20%

highlighting the advantages of each module in modeling complex temporal patterns. Additional validations in cross-regional （wind farms in Central and Western China） and cross-energy （photovoltaic plants） scenarios further demonstrate the model's strong generalization capability.

关键词

Keywords

references

刘德澳, 董明刚, 叶威, 等. 基于图神经网络的非均匀风电场功率预测方法[J]. 太阳能学报, 2023, 44(11): 279-286.

荆志宇, 李培强, 林文婷. 结合贝叶斯优化及通道注意力的双端优化时序式风电功率预测模型[J]. 电力系统及其自动化学报, 2024, 36(8): 39-47, 59.

唐清苇, 向月, 代佳琨, 等. 基于CNN-LSTM的风电场发电功率迁移预测方法[J]. 工程科学与技术, 2024, 56(2): 91-99.

琚垚, 祁林, 刘帅. 基于改进乌鸦算法和ESN神经网络的短期风电功率预测[J]. 电力系统保护与控制, 2019, 47(4): 58-64.

李润金, 李丽霞. 基于LSTM神经网络的风电功率预测研究[J]. 沈阳工程学院学报(自然科学版), 2023, 19(3): 14-18.

薛阳, 王琳, 王舒, 等. 一种结合CNN和GRU网络的超短期风电预测模型[J]. 可再生能源, 2019, 37(3): 456-462.

郭凯杰, 耿光超, 江全元, 等. 计及超短期预测误差的风储系统跟踪计划出力控制策略[J]. 太阳能学报, 2023, 44(2): 326-333.

李宏扬, 高丙朋. 基于改进VMD和SNS-Attention-GRU的短期光伏发电功率预测[J]. 太阳能学报, 2023, 44(8): 292-300.

王献志, 曾四鸣, 周雪青, 等. 基于XGBoost联合模型的光伏发电功率预测[J]. 太阳能学报, 2022, 43(4): 236-242.

胡云峰. 基于短期风电功率预测的桨距角优化研究[D].吉林:东北电力大学,2023,1-64.

钱勇生, 邵洁, 季欣欣, 等. 基于LSTM-Attention网络的短期风电功率预测[J]. 电机与控制应用, 2019, 46(9): 95-100.

杨茂, 黄宾阳, 江博, 等. 基于卡尔曼滤波和支持向量机的风电功率实时预测研究[J]. 东北电力大学学报, 2017, 37(2): 45-51.

黄万超. 机器学习技术在风力与光伏发电系统短期功率预测中的应用与比较研究[J]. 现代工业经济和信息化, 2024, 14(10): 137-138.

ZHOU Y, RAVEY A, PERA M C.Real-time predictive energy management for fuel cell electric vehicles[C]//2021 IEEE Transportation Electrification Conference & Expo (ITEC). Chicago, IL, USA, 2021: 142-147.

SHANG Z H, CHEN Y H, LAI D K, et al.A novel interpretability machine learning model for wind speed forecasting based on feature and sub-model selection[J]. Expert systems with applications, 2024, 255: 124560.

XU H Y, CHANG Y Q, ZHAO Y, et al.A hybrid model for multi-step wind speed forecasting based on secondary decomposition, deep learning, and error correction algorithms[J]. Journal of intelligent & fuzzy systems, 2021, 41(2): 3443-3462.

BLFGEH A, ALKHUDHAYR H.A machine learning-based sustainable energy management of wind farms using Bayesian recurrent neural network[J]. Sustainability, 2024, 16(19): 8426.

KHODAYAR M, KAYNAK O, KHODAYAR M E.Rough deep neural architecture for short-term wind speed forecasting[J]. IEEE transactions on industrial informatics, 2017, 13(6): 2770-2779.

NIKSA-RYNKIEWICZ T, STOMMA P, WITKOWSKA A, et al.An intelligent approach to short-term wind power prediction using deep neural networks[J]. Journal of artificial intelligence and soft computing research, 2023, 13(3): 197-210.

CHEN Y B, XU J J.Solar and wind power data from the Chinese State Grid Renewable Energy Generation Forecasting Competition[J]. Scientific data, 2022, 9: 577.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

基于ICPO优化VMD耦合深度学习模型的中短期风电功率预测

基于VMD和多时间尺度分类预测的风电单机短期功率预测研究

基于多通道特征融合优化VMD-SCNN-LSTM的锂电池RUL预测

基于VMD-TCN-GRU-AM的超短期风电预测

PHOTOVOLTAIC HOT SPOT RECOGNITION BASED ON ATTENTION MECHANISM

Related Author

黄伟

刘彬

李火坤

黄俊

黄梓阳

夏卫平

邓艾东

薛原

Related Institution

Department of Automation, North China Electric Power University

Hebei Technology Innovation Center of Simulation ＆ Optimized Control for Power Generation, North China Electric Power University

School of Economics and Management, North China Electric Power University

Economic and Technology Research Institute, State Grid Chongqing Electric Power Company

School of Electrical Engineering, Xinjiang University

AI问答

⁰