Deep temporal convolutional networks for F10.7 radiation flux short-term forecasting

IF 1.7 4区 地球科学 Q3 ASTRONOMY & ASTROPHYSICS Annales Geophysicae Pub Date : 2024-04-12 DOI:10.5194/angeo-42-91-2024
Luyao Wang, Hua Zhang, Xiaoxin Zhang, Guangshuai Peng, Zheng Li, Xiaojun Xu
{"title":"Deep temporal convolutional networks for F10.7 radiation flux short-term forecasting","authors":"Luyao Wang, Hua Zhang, Xiaoxin Zhang, Guangshuai Peng, Zheng Li, Xiaojun Xu","doi":"10.5194/angeo-42-91-2024","DOIUrl":null,"url":null,"abstract":"Abstract. F10.7, the solar flux at a wavelength of 10.7 cm (F10.7), is often used as an important parameter input in various space weather models and is also a key parameter for measuring the strength of solar activity levels. Therefore, it is valuable to study and forecast F10.7. In this paper, the temporal convolutional network (TCN) approach in deep learning is used to predict the daily value of F10.7. The F10.7 series from 1957 to 2019 are used. The data during 1957–1995 are adopted as the training dataset, the data during 1996–2008 (solar cycle 23) are adopted as the validation dataset, and the data during 2009–2019 (solar cycle 24) are adopted as the test dataset. The leave-one-out method is used to group the dataset for multiple validations. The prediction results for 1–3 d ahead during solar cycle 24 have a high correlation coefficient (R) of 0.98 and a root mean square error (RMSE) of only 5.04–5.18 sfu. The overall accuracy of the TCN forecasts is better than the autoregressive (AR) model (it only takes past values of the F10.7 index as inputs) and the results of the US Space Weather Prediction Center (SWPC) forecasts, especially for 2 and 3 d ahead. In addition, the TCN model is slightly better than other neural network models like the backpropagation (BP) neural network and long short-term memory (LSTM) network in terms of the solar radiation flux F10.7 forecast. The TCN model predicted F10.7 with a lower root mean square error, a higher correlation coefficient, and a better overall model prediction.","PeriodicalId":50777,"journal":{"name":"Annales Geophysicae","volume":null,"pages":null},"PeriodicalIF":1.7000,"publicationDate":"2024-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annales Geophysicae","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.5194/angeo-42-91-2024","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
引用次数: 0

Abstract

Abstract. F10.7, the solar flux at a wavelength of 10.7 cm (F10.7), is often used as an important parameter input in various space weather models and is also a key parameter for measuring the strength of solar activity levels. Therefore, it is valuable to study and forecast F10.7. In this paper, the temporal convolutional network (TCN) approach in deep learning is used to predict the daily value of F10.7. The F10.7 series from 1957 to 2019 are used. The data during 1957–1995 are adopted as the training dataset, the data during 1996–2008 (solar cycle 23) are adopted as the validation dataset, and the data during 2009–2019 (solar cycle 24) are adopted as the test dataset. The leave-one-out method is used to group the dataset for multiple validations. The prediction results for 1–3 d ahead during solar cycle 24 have a high correlation coefficient (R) of 0.98 and a root mean square error (RMSE) of only 5.04–5.18 sfu. The overall accuracy of the TCN forecasts is better than the autoregressive (AR) model (it only takes past values of the F10.7 index as inputs) and the results of the US Space Weather Prediction Center (SWPC) forecasts, especially for 2 and 3 d ahead. In addition, the TCN model is slightly better than other neural network models like the backpropagation (BP) neural network and long short-term memory (LSTM) network in terms of the solar radiation flux F10.7 forecast. The TCN model predicted F10.7 with a lower root mean square error, a higher correlation coefficient, and a better overall model prediction.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于 F10.7 辐射通量短期预报的深度时空卷积网络
摘要F10.7,即波长为 10.7 厘米(F10.7)的太阳通量,经常被用作各种空间天气模式的重要输入参数,也是测量太阳活动强度的关键参数。因此,研究和预测 F10.7 具有重要价值。本文采用深度学习中的时序卷积网络(TCN)方法来预测 F10.7 的日值。本文使用 1957 年至 2019 年的 F10.7 系列数据。1957-1995 年的数据作为训练数据集,1996-2008 年(太阳周期 23)的数据作为验证数据集,2009-2019 年(太阳周期 24)的数据作为测试数据集。采用 "留一弃一 "的方法对数据集进行分组,以进行多次验证。太阳周期 24 期间提前 1-3 d 的预测结果相关系数(R)高达 0.98,均方根误差(RMSE)仅为 5.04-5.18 sfu。TCN预报的整体准确性优于自回归(AR)模式(它只将过去的F10.7指数值作为输入)和美国空间天气预报中心(SWPC)的预报结果,尤其是在未来2天和3天。此外,在太阳辐射通量 F10.7 预报方面,TCN 模型略优于其他神经网络模型,如反向传播(BP)神经网络和长短期记忆(LSTM)网络。TCN 模型预测 F10.7 的均方根误差更小、相关系数更高,而且模型的整体预测效果更好。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Annales Geophysicae
Annales Geophysicae 地学-地球科学综合
CiteScore
4.30
自引率
0.00%
发文量
42
审稿时长
2 months
期刊介绍: Annales Geophysicae (ANGEO) is a not-for-profit international multi- and inter-disciplinary scientific open-access journal in the field of solar–terrestrial and planetary sciences. ANGEO publishes original articles and short communications (letters) on research of the Sun–Earth system, including the science of space weather, solar–terrestrial plasma physics, the Earth''s ionosphere and atmosphere, the magnetosphere, and the study of planets and planetary systems, the interaction between the different spheres of a planet, and the interaction across the planetary system. Topics range from space weathering, planetary magnetic field, and planetary interior and surface dynamics to the formation and evolution of planetary systems.
期刊最新文献
Ionospheric upwelling and the level of associated noise at solar minimum Sensitivity analysis of a Martian atmospheric column model with data from the Mars Science Laboratory Low-frequency solar radio type II bursts and their association with space weather events during the ascending phase of solar cycle 25 The investigation of June 21 and 25, 2015 CMEs using EUHFORIA Observations of ionospheric disturbances associated with the 2020 Beirut explosion by Defense Meteorological Satellite Program and ground-based ionosondes
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1