空间科学学报2025，Vol.45Issue(5)：1230-1242,13.DOI:10.11728/cjss2025.05.2024-0087
四种神经网络方法在电离层磁暴期TEC建模的对比分析
Comparative Analysis of Four Neural Network Methods for TEC Modeling during Ionospheric Magnetic Storms
摘要
Abstract
The Total Electron Content(TEC)of the ionosphere is an important parameter for describ-ing the ionosphere activities,and much research has been done for the accurate methods for the iono-spheric TEC prediction.However,the prediction accuracy of ionospheric empirical models for TEC dur-ing geomagnetic storms is still not ideal.To address this issue,this paper aims to assess the performance of ionospheric TEC predicting methods,which involve the LSTM,the BiLSTM,the Convolutional Neu-ral Network-Long Short-Term Memory combined with Attention mechanism(CNN-LSTM-Attention),and the Convolutional Neural Network-Bidirectional Long Short-Term Memory combined with Atten-tion mechanism(CNN-BiLSTM-Attention).At first,the geomagnetic storm periods are identified by comparing with the threshold of Dst index(≤-30 nT),during the years from 2004 to 2022.Then,four neural network models for the ionospheric TEC prediction are formed,through the combinations of mul-tiple spatiotemporal parameters,such as UTS,UTC,SA,AA,CHS,and SHS.Finally,the accuracy and reliability of the four neural network models are assessed using the reference TEC dataset collected dur-ing geomagnetic storms in 2023,and three statistical index,Mean Absolute Error(MAE),The Root Mean Square Error(RMSE),and coefficient of determination R2,are utilized.The results show that,the performance of the CNN-BiLSTM-Attention model is superior to the other three models,with MAE ranging from 0.882 to 5.270 TECU,RMSE between 1.175 and 6.983 TECU,and R2 values exceeding 0.7.In order to better describe the difference between the predicted values and the reference values,the scat-ter plots of two datasets are plotted for the fitting of linear regression equations.The slope of fitted func-tion from CNN-BiLSTM-Attention model is very close to the ideal value 1,also indicating a better per-formance compared to the other models.关键词
电离层磁暴期/电离层总电子含量/长短期记忆网络/卷积神经网络/注意力机制Key words
Ionospheric magnetic storm period/Total electron content of ionosphere/Long short-term memory/Convolutional neural network/Attention mechanism分类
地球科学引用本文复制引用
朱佳豪,闫文林,金宇峰,严泰明,王坚..四种神经网络方法在电离层磁暴期TEC建模的对比分析[J].空间科学学报,2025,45(5):1230-1242,13.基金项目
国家自然科学基金项目(42274029)和江苏省自然科学基金项目(BK20181015)共同资助 （42274029）
