Application of Improved Model Based on LSTM in Ionospheric TEC Forecast

电离层延迟是全球卫星导航定位中重要的误差源之一, 提高电离层总电子含量(Total Electron Content, TEC)预报精度对改善卫星导航定位精度极其重要. 本文联合滑动窗口(Sliding Window)和长短时记忆(Long Short-Term Memory, LSTM))神经网络, 以滑动窗口算法对输入序列数据集不断更新并测试不同输入序列长度对应模型的精度, 最后以预测值来更新输入数据序列的最后10%, 进而构建TEC预报模型. 验证结果表明, 该模型在平静期和磁暴期预测残差绝对值小于5 TECU的比例均达85%以上, 较传统LSTM模型对应值占比增加了49%, 71%, 均方根误差(RMSE)低31%, 35%; 其预报结果的平均绝对误差(MAE)减少25%, 32%; SLSTM模型预测结果的RMSE均值、MAE均值均比传统LSTM模型、BP模型小.

Authors

Publication Details

Journal
Chinese Journal of Space Science
Published
2025-01-01
DOI
https://doi.org/10.11728/cjss2025.05.2024-0112
Primary Topic
Earthquake Detection and Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Application of Improved Model Based on LSTM in Ionospheric TEC Forecast

Junyu LI, Lin Wei, Liangke Huang, Can Huang et al.
Chinese Journal of Space Science
Earthquake Detection and Analysis
article

Application of Improved Model Based on LSTM in Ionospheric TEC Forecast

Junyu LI, Lin Wei, Liangke Huang, Can Huang, Lilong Liu
article en

Abstract

电离层延迟是全球卫星导航定位中重要的误差源之一, 提高电离层总电子含量(Total Electron Content, TEC)预报精度对改善卫星导航定位精度极其重要. 本文联合滑动窗口(Sliding Window)和长短时记忆(Long Short-Term Memory, LSTM))神经网络, 以滑动窗口算法对输入序列数据集不断更新并测试不同输入序列长度对应模型的精度, 最后以预测值来更新输入数据序列的最后10%, 进而构建TEC预报模型. 验证结果表明, 该模型在平静期和磁暴期预测残差绝对值小于5 TECU的比例均达85%以上, 较传统LSTM模型对应值占比增加了49%, 71%, 均方根误差(RMSE)低31%, 35%; 其预报结果的平均绝对误差(MAE)减少25%, 32%; SLSTM模型预测结果的RMSE均值、MAE均值均比传统LSTM模型、BP模型小.

Chinese Journal of Space ScienceVol. 45(5)
Openalex Percentile: Top 73%
Earthquake Detection and Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.