改进的GM-AR组合模型在电离层TEC预报中的应用

电离层总电子含量(TEC)是影响GPS导航定位精度的重要因素之一。因此对于电离层总电子含量的研究及准确预测可以大大提高GPS的定位精度。由于灰色预测模型在理论上可以进行中长期预测,但实际应用中随着时间的推移,其预测精度会随之下降,为解决这一问题,对GM(1,1)模型进行改进,并将改进后的GM(1,1)模型与时间序列模型组合。利用改进的GM-AR模型进行TEC预报,预报的精度比两种方法单独预测的精度有较大提高,并应用实例证明该方法的可行性。

Application of GM-AR model to the prediction of ionospheric TEC

The ionospheric Total Electron Content (TEC) is one of the important factors affecting the accuracy of GPS navigation and positioning .So the study on the ionospheric Total Electron Content and accurate prediction of GPS positioning accuracy can be greatly improved .The gray prediction model can serve for long term prediction in theory ,but in practice with the passage of time ,the prediction accuracy will drop .In order to solve this problem ,GM (1 ,1) model is improved ,and combines the improved GM (1 ,1) model with time sequence .Using improved GM‐AR model for TEC forecast ,its accuracy is more than the two methods separately ,which its application example proves feasible .