基于实时天气预报尺度风场的精细化台风三维风场重构方法研究

刘皓月; 任贺贺; 柯世堂; 邱嘉琦

doi:doi:10.19912/j.0254-0096.tynxb.2024-1743

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 >

基于实时天气预报尺度风场的精细化台风三维风场重构方法研究

更新时间：2026-03-06

- 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究
- 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究
- 太阳能学报 2026年47卷第2期页码：462-469
- 作者机构：
- 作者简介：
- 基金信息：
  
  国家自然科学基金（52478530;52321165649）;博士后自然科学基金（2022M711618）;中央高校基本科研业务费专项资金（NC2024006）
- DOI：doi:10.19912/j.0254-0096.tynxb.2024-1743
  中图分类号： TU312
- 网络首发：2026-03-05，
  
  纸质出版：2026-03-05
- 稿件说明：
移动端阅览
刘皓月, 任贺贺, 柯世堂, 等. 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究[J]. 太阳能学报, 2026,47(2):462-469.

刘皓月, 任贺贺, 柯世堂, et al. 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究[J]. 2026, 47(2): 462-469.
刘皓月, 任贺贺, 柯世堂, 等. 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究[J]. 太阳能学报, 2026,47(2):462-469. DOI： doi:10.19912/j.0254-0096.tynxb.2024-1743.

刘皓月, 任贺贺, 柯世堂, et al. 基于实时天气预报尺度风场的精细化台风三维风场重构方法研究[J]. 2026, 47(2): 462-469. DOI： doi:10.19912/j.0254-0096.tynxb.2024-1743.

摘要

基于机器学习中的超分辨率重构技术

提出基于混合下采样跳跃连接（DSC）/多尺度（MS）模型的台风三维风场超分辨率重构方法

开展不同网格尺度之间的三维风场超分辨率重构研究。通过对比分析台风整体流场、空间不同区域平均风剖面、径向风速分布等误差

结果表明

千米及以下尺度风场可较好地重构精细化台风三维风场

而千米尺度以上风场重构误差较大

初步实现了台风风场重构从千米量级到百米量级跨越。

Abstract

Based on the super-resolution reconstruction technique in machine learning

this paper proposes a hybrid downsampling skip connection （DSC）/multi-scale （MS） model for super-resolution reconstruction of three-dimensional typhoon wind fields

and conducts research on super-resolution reconstruction of three-dimensional wind fields across different grid scales. Through comparative analysis of errors in the overall typhoon flow field

average wind profiles in different spatial regions

and radial wind speed distributions

the results indicate that wind fields at kilometer-scale and below can effectively reconstruct refined three-dimensional typhoon wind fields

while reconstruction errors are relatively larger for wind fields above kilometer-scale. This study preliminarily achieves a transition in typhoon wind field reconstruction from kilometer-scale to hectometer-scale.

关键词

Keywords

references

韩然, 王珑, 王同光, 等. 台风不同区域中的风力机动力响应特性研究[J]. 太阳能学报, 2020, 41(10): 251-258.

史军, 徐家良, 穆海振. 上海近海海上最大风速的估算及数值模拟[J]. 太阳能学报, 2017, 38(4): 991-998.

李强, 张秀芝, 王乔乔, 等. 15°~35°N、105°~130°E海域台风极值风速分析[J]. 太阳能学报, 2015, 36(1): 85-89.

SPIRIDONOV V, BAEZ J, TELENTA B, et al.Prediction of extreme convective rainfall intensities using a free-running 3-D sub-km-scale cloud model initialized from WRF km-scale NWP forecasts[J]. Journal of atmospheric and solar-terrestrial physics, 2020, 209: 105401.

余文林, 柯世堂. 基于WRF与CFD嵌套的台风下大型风力机流场作用与气动力分布[J]. 太阳能学报, 2020, 41(12): 260-269.

BRUNTON S L, NOACK B R.Closed-loop turbulence control: progress and challenges[J]. Applied mechanics reviews, 2015, 67(5): 050801.

SRINIVASAN P A, GUASTONI L, AZIZPOUR H, et al.Predictions of turbulent shear flows using deep neural networks[J]. Physical review fluids, 2019, 4(5): 054603.

AVANZO M, GAGLIARDI V, STANCANELLO J, et al.Combining computed tomography and biologically effective dose in radiomics and deep learning improves prediction of tumor response to robotic lung stereotactic body radiation therapy[J]. Medical physics, 2021, 48(10): 6257-6269.

SELVI S, AGGARWAL K, PANDURANGAN R, et al.Retraction Note: enhancing the accuracy of target detection in remote video surveillance analytics through federated learning[J]. Optical and quantum electronics, 2024, 56(10): 1769.

KIM J, LEE J K, LEE K M.Accurate image super-resolution using very deep convolutional networks[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA, 2016: 1646-1654.

LAW H, DENG J.CornerNet: detecting objects as paired keypoints[C]//Computer Vision-ECCV 2018. Cham: Springer, 2018: 765-781.

WU K, LI X M.Deep learning for retrieving omni-directional ocean wave spectra from spaceborne synthetic aperture radar[J]. Remote sensing of environment, 2024, 314: 114386.

FUKAMI K, FUKAGATA K, TAIRA K.Super-resolution reconstruction of turbulent flows with machine learning[J]. Journal of fluid mechanics, 2019, 870: 106-120.

FUKAMI K, FUKAGATA K, TAIRA K.Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows[J]. Journal of fluid mechanics, 2021, 909: A9.

KIM H, KIM J, WON S, et al.Unsupervised deep learning for super-resolution reconstruction of turbulence[J]. Journal of fluid mechanics, 2021, 910: A29.

ZHANG J C, ZHAO X W.Three-dimensional spatiotemporal wind field reconstruction based on physics-informed deep learning[J]. Applied energy, 2021, 300: 117390.

FARANDA D, MESSORI G, BOURDIN S, et al.Correcting biases in tropical cyclone intensities in low-resolution datasets using dynamical systems metrics[J]. Climate dynamics, 2023, 61(9): 4393-4409.

战庆亮, 刘鑫, 张冠华, 等. 桥面风场时程重构的机器学习方法[J]. 中国公路学报, 2023, 36(8): 22-31.

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.

李聪健, 高航, 刘奕. 基于数值模拟和机器学习的风场快速重构方法[J]. 清华大学学报(自然科学版), 2023, 63(6): 882-887.

NGIAM J, CHEN Z, CHIA D, et al.Tiled convolutional neural networks[J]. Advances in neural information processing systems, 2010, 23(1): 1279-1287.

HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA, 2016: 770-778.

DU X F, QU X B, HE Y F, et al.Single image super-resolution based on multi-scale competitive convolutional neural network[J]. Sensors, 2018, 18(3): 789.

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

台风期间福建省平潭近海近地层风特性研究

台风过境全过程水平风速变化中海气交换作用机制与影响因子研究

台风环境下考虑海洋运动的台湾海峡风能波浪能联合相关性分析

深度强化学习优化非线性自抗扰微网稳压技术

考虑碳排放责任的多微网配电系统混合博弈低碳策略