Research Of Short Term Prediction Model Of Total Electron Content In Ionospheric

The ionosphere is an important part of the solar terrestrial space environment,which has an important impact on radio communication,navigation,satellite positioning and human space activities.As an important parameter to describe the characteristics of the ionosphere,the prediction and analysis of the total electron content(TEC)has always been the focus of ionospheric research.According to the characteristics of Ionospheric TEC data,such as nonlinearity and high noise,this paper introduces two methods to preprocess the data,so as to establish the Ionospheric TEC short-term prediction model..The main research contents are as follows(1)In view of the nonlinear and nonstationary characteristics of Ionospheric TEC data,empirical wavelet transform(EWT)is applied to the short-term prediction of Ionospheric TEC.Based on the idea of decomposition prediction reconstruction,EWT-ARMA combined model is established to predict TEC data in different solar activity years,and EWT-Elman Neural Network prediction model is established to predict TEC data in different geomagnetic environments.The results show that EWT can improve the Ionospheric TEC prediction accuracy of the two prediction models in different environments.(2)According to the characteristics of Ionospheric TEC data discreteness and disorder,this paper introduces the Prophet model to preprocess the Ionospheric TEC data,and carries out the model prediction for the fitting data and residual data,so as to improve the prediction accuracy.First of all,the Prophet-ARMA residual correction model is established to forecast and analyze the TEC data of the same time period and different time periods of the same position provided by IGS center in 2010.At the same time,the Prophet-Elman residual correction model is established to forecast and analyze the TEC data of different solar activity years.The results show that the application of the Prophet model to the Elman Neural Network in the Ionospheric TEC forecast is feasible,It can effectively improve the prediction accuracy of TEC.(3)In view of the influence of solar activity and geomagnetic environment on Ionospheric TEC,the EWT-Elman combined forecasting model and the Prophet-Elman residual correction model are used to forecast the Ionospheric TEC data with different solar activity,geomagnetic environment,latitude and seasonal variation,and the results are compared with those of ARMA model and Elman model.The results show that EWT Elman model and Prophet-Elman residual correction model have better prediction effect in different environments,and have their own advantages in different environments.