电离层总电子含量(Total Electron Content,TEC)的预测是空间环境研究的重要内容,对卫星通讯和导航定位等有重要意义.设计了一个包含注意力机制的LSTM预测模型,利用连续5天的TEC来预测未来1天的TEC.在两个站点上将本文提出的模型与ARIMA和LSTM进行了对比实验.结果表明,所提模型的拟合程度可达0.9965,明显优于对比模型.
Prediction of Ionospheric Total Electron Content Based on Attention Mechanism LSTM
The prediction of Total Electron Content(TEC)in the ionosphere is an important content of space environment research,which is of great significance to satellite communication,navigation and positioning.In this paper,an LSTM prediction model in-cluding attention mechanism is designed,and the TEC for five consecutive days is used to predict the TEC for the next day.In this paper,the model proposed is compared with ARIMA and LSTM in two sites.The results show that the fitting degree of the model proposed can reach 0.9965,which is obviously better than the contrast model.
NETL
NSTL
万方数据
