Research And Application Of Prediction Model Of Total Electron Content In Ionosphere

The ionosphere is an important component of the atmosphere between the earth and the sun.Total Electron Content(TEC)is an important parameter to characterize the spatio-temporal variations of ionosphere.The study of ionospheric TEC can provide theory and data foundation to delay correction of radio signals and prediction of short-term earthquake.This paper has done the following work:First of all,describing the contents and format of TEC data provided by IGS in detail.Using the TEC data from 2000 and 2009 provided by IGS to analyze the effects of solar activity,geomagnetic activity,diurnal variation and seasonal variation on ionospheric TEC.The experimental results show that TEC changes are positively correlated with solar activity.Magnet storm disperses the ionosphere.The ionospheric TEC diurnal changes first from west to east,and then gradually from east to west.However,the seasonal performance of TEC is firstly from north to south,then changing from south to north.Then,introducing the Empirical Mode Decomposition(EMD),Ensemble Empirical Mode Decomposition(EEMD)and Complementary Ensemble Empirical Mode Decomposition(CEEMD).Analyzing their deficiencies and introduce the improved Modified Ensemble Empirical Mode Decomposition(MEEMD).Using the EMD,EEMD,CEEMD,MEEMD and improved MEEMD to decompose the simulated signal,and comparing the time-frequency spectra of MEEMD and improved MEEMD.Then performming the decomposition of ionospheric TEC data by MEEMD and improved MEEMD,and And analyze the spectrums of the components.Simultaneously,proposing an improved MEEMD-ARMA residuals correction model and introducing its principle in detail.extracting the ionospheric TEC data of the same position and different periods of time,the same longitude and period of time but different latitudes and the same latitude and period of time but longitudes of 2000 and 2009.Based on the TEC data of the first 20 days,respectively using the ARMA model and the improved MEEMD-ARMA model to predict the TEC data of the last 8 days,and analyzing the forecast results in detail.Finally,putting forward the improved MEEMD-ARMA residuals corroction model of seismic ionospheric TEC detection,and comparing the prediction accuracy of the ionospheric TEC reference background values of this method and the traditional detection methods,it is found that percentage of the absolute residuals less than 3 TECU of the predicted background values of the model of this paper is 91.67%,higher than the traditional models more than 6.25 percentage points.Introducing the concrete implementation method of the model of this paper about background value calculation in the earthquake ionospheric TEC detection,determining the twice standard deviation of absolute residuals of the predicted reference background values as the detection limit,and probing the Lushan earthquake on April 20,2013.After excluding the influence of solar and geomagnetic activities,it is found that the obvious positive anomalies at 9 days and 13 days before the earthquake and the obvious negative anomaly at 4 days before the earthquake may be caused by the earthquake.