一种风矢量分解和RobustSTL-TimesNet-BiGRU的复杂地形风向预测

刘洋; 王聪; 张宏立; 马萍; 李新凯

doi:10.19912/j.0254-0096.tynxb.2023-1828

欢迎来到能源电力期刊网！

English Version 读者登录期刊登录

您当前的位置：

首页 >

文章列表页 >

一种风矢量分解和RobustSTL-TimesNet-BiGRU的复杂地形风向预测

更新时间：2025-10-18

- 一种风矢量分解和RobustSTL-TimesNet-BiGRU的复杂地形风向预测
- WIND VECTOR DECOMPOSITION AND ROBUSTSTL-TIMESNET-BIGRU WIND DIRECTION FORECASTING IN COMPLEX TERRAIN
- 太阳能学报 2025年46卷第3期页码：576-588
- 作者机构：
  
  1. 新疆大学电气工程学院,乌鲁木齐 830047
  2. 智能科学与技术学院,新疆大学工程训练中心,乌鲁木齐 830047
- 作者简介：
- 基金信息：
  
  国家重点研发计划（2021YFB1507000）;新疆维吾尔自治区自然科学基金（2022D01E33;2022D01C367）;国家自然科学基金（52267010）;自治区重点研发计划（2022B03031）
- DOI：10.19912/j.0254-0096.tynxb.2023-1828
  中图分类号： TM614
- 网络出版：2025-04-07，
  
  纸质出版：2025
- 稿件说明：
移动端阅览
刘洋, 王聪, 张宏立, 马萍, 李新凯. 一种风矢量分解和RobustSTL-TimesNet-BiGRU的复杂地形风向预测[J]. 太阳能学报, 2025, 46(3): 576-588.

Liu Yang, Wang Cong, Zhang Hongli, et al. WIND VECTOR DECOMPOSITION AND ROBUSTSTL-TIMESNET-BIGRU WIND DIRECTION FORECASTING IN COMPLEX TERRAIN[J]. 2025, 46(3): 576-588.
刘洋, 王聪, 张宏立, 马萍, 李新凯. 一种风矢量分解和RobustSTL-TimesNet-BiGRU的复杂地形风向预测[J]. 太阳能学报, 2025, 46(3): 576-588. DOI： 10.19912/j.0254-0096.tynxb.2023-1828.

Liu Yang, Wang Cong, Zhang Hongli, et al. WIND VECTOR DECOMPOSITION AND ROBUSTSTL-TIMESNET-BIGRU WIND DIRECTION FORECASTING IN COMPLEX TERRAIN[J]. 2025, 46(3): 576-588. DOI： 10.19912/j.0254-0096.tynxb.2023-1828.

摘要

针对复杂地形下的风向预测场景

提出一种风矢量正交分解、鲁棒性局部加权回归下的周期趋势分解（RobustSTL）方法、TimesNet模型和融合双向门控循环单元网络（BiGRU）误差补偿的多步风向预测方法。首先

为了减少原始风向循环圆周特性带来的大幅度波动性

将风向与相关性强的风速利用风矢量正交分解方法转化为波动性较小的矢量风速

并利用RobustSTL将矢量风速分解为趋势项、季节项和剩余波动项。其次

将分解后的各项分别训练TimesNet网络并得到各项的初步预测结果

对各项进行求和并重构为初始预测风向。然后

为了进一步挖掘初步风向预测误差的深层特征

提高风向的预测精度

采用BiGRU对初步预测误差进行建模与训练。最后

将预测的误差与初步预测风向加和

得到最终的风向预测结果。采用实际复杂地形风电场数据进行验证分析

结果表明所提的多步风向预测混合模型具有较高的预测精度。

Abstract

Aiming at wind direction forecasting in complex terrain

this paper proposes a multi-step wind direction forecasting method based on wind vector orthogonal decomposition

Robust Seasonal-Trend decomposition using Loess （RobustSTL）

the TimesNet model

and a bidirectional gated recurrent unit network （BiGRU） for error compensation. Firstly

to reduce the significant fluctuation caused by the circular characteristics of the original wind direction

the wind direction and the correlated wind speed are converted into a less fluctuating vector wind speed using the wind vector orthogonal decomposition method. This vector wind speed is then decomposed into trend

seasonal

and residual components using RobustSTL. Secondly

each component after decomposition is trained using the TimesNet model

and preliminary forecasting results are obtained. The components are then summed and reconstructed to form the initial wind direction forecast. To further extract deep features from the initial forecasting errors and enhance accuracy

the BiGRU network is used for modeling and training the forecasting errors. Finally

the predicted error is combined with the initial forecast to obtain the final wind direction forecast. The proposed multi-step wind direction forecasting hybrid model is validated using real data from a complex terrain wind farm

and the results demonstrate its high forecasting accuracy.

关键词

Keywords

references

WANG Q, LUO K, WU C L, et al.Mesoscale simulations of a real onshore wind power base in complex terrain: wind farm wake behavior and power production[J]. Energy, 2022, 241: 122873.

朱蓉, 向洋, 孙朝阳, 等. 中国典型复杂地形风能资源特性及其形成机制[J]. 太阳能学报, 2024, 45(4): 226-237.

ERDEM E, SHI J.ARMA based approaches for forecasting the tuple of wind speed and direction[J]. Applied energy, 2011, 88(4): 1405-1414.

KAVASSERI R G, SEETHARAMAN K.Day-ahead wind speed forecasting using f-ARIMA models[J]. Renewable energy, 2009, 34(5): 1388-1393.

KHOSRAVI A, KOURY R N N, MACHADO L, et al. Prediction of wind speed and wind direction using artificial neural network, support vector regression and adaptive neuro-fuzzy inference system[J]. Sustainable energy technologies and assessments, 2018, 25: 146-160.

LANG M N, SCHLOSSER L, HOTHORN T, et al.Circular regression trees and forests with an application to probabilistic wind direction forecasting[J]. Journal of the royal statistical society series C: applied statistics, 2020, 69(5): 1357-1374.

TAGLIAFERRI F, VIOLA I M, FLAY R G J. Wind direction forecasting with artificial neural networks and support vector machines[J]. Ocean engineering, 2015, 97: 65-73.

唐振浩, 赵赓楠, 曹生现, 等. 一种基于数据解析的混合风向预测算法[J]. 太阳能学报, 2021, 42(9): 349-356.

朱天宇, 叶强, 郝建树, 等. 基于LSTM的风矢量预测方法[J]. 电力自动化设备, 2023, 43(11): 111-116.

CHITSAZAN M A, SAMI FADALI M, TRZYNADLOWSKI A M.Wind speed and wind direction forecasting using echo state network with nonlinear functions[J]. Renewable energy, 2019, 131: 879-889.

DING Y, YE X W, GUO Y.A multistep direct and indirect strategy for predicting wind direction based on the EMD-LSTM model[J]. Structural control and health monitoring, 2023, 2023: 4950487.

陈臣鹏, 赵鑫, 毕贵红, 等. 基于多模式分解和麻雀优化残差网络的短期风速预测模型[J]. 电网技术, 2022, 46(8): 2975-2985.

王颖, 朱南阳, 谢浩川, 等. 基于对比学习辅助训练的超短期风功率预测方法[J]. 仪器仪表学报, 2023, 44(3): 89-97.

VASWANI A, SHAZEER N, Parmar N, et al.Attention is all you need[J]. Advances in neural information processing systems, 2017, 30: 5998-6008.

吉兴全, 曾若梅, 张玉敏, 等. 基于注意力机制的CNN-LSTM短期电价预测[J]. 电力系统保护与控制, 2022, 50(17): 125-132.

SARI A P, SUZUKI H, KITAJIMA T, et al.Short-term wind speed and direction forecasting by3DCNNand deep ConvolutionalLSTM[J]. IEEJ transactions on electrical and electronic engineering, 2022, 17(11): 1620-1628.

CLEVELAND R B, CLEVELAND W S, MCRAE J E, et al.STL: A seasonal-trend decomposition[J]. Journal of Official Statistics, 1990, 6(1): 3-73.

WEN Q S, GAO J K, SONG X M, et al.RobustSTL: a robust seasonal-trend decomposition algorithm for long time series[J]. Proceedings of the AAAI conference on artificial intelligence, 2019, 33(1): 5409-5416.

王瑞, 徐新超, 逯静. 基于特征选择及ISSA-CNN-BiGRU的短期风功率预测[J]. 工程科学与技术, 2024, 56(3): 228-239.

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MATLAB系统辨识工具箱的风信号预测

基于风电机组聚类的风电场有功分层分配策略