小波分解层数对电离层预报的影响

针对电离层总电子含量(TEC)所具有的非线性、非平稳的特性,本文将小波分析理论引入到时间序列分析方法中,对IGS中心提供的不同纬度前8 d电离层格网点观测数据进行不同层数的分解,而后采用ARIMA模型预报后4d TEC数据。将预报结果与观测数据对比并统计精度,实验结果表明对电离层格网点数据进行1层~3层分解能够取得较好的预报效果,并且采用1层分解所得预报效果最佳。

Wavelet Decomposition Level Impact on Ionospheric Prediction

This paper brings wavelet analysis theory into time series analysis method aiming at the nonlinear and non-stationary characteristic of Total Electronic Content in ionosphere. We decomposed the former 8 days’ ionospheric grid observe TEC data in different latitude provided by IGS center in different layer,then forecast the latter 4 days’ TEC data with ARIMA model. Then we compare the forecast result with observe data and get the statistical accuracy, the result shows that we can achieve a rather good prediction effect decomposed in 1~3 layers. Also,we can get the best forecast result just decomposed in 1 layer.