期刊文献+

基于CEEMD的电离层TEC组合预报模型

Integrated prediction model of ionospheric TEC based on CEEMD
下载PDF
导出
摘要 针对单一电离层总电子含量(TEC)预报模型存在的缺陷,如受外界因素干扰较大、预报精度随预报时间的增加明显降低等,本文提出一种基于补充集合经验模态分解(CEEMD)电离层TEC组合预报模型。该模型实现电离层TEC预报的关键途径为:首先,利用CEEMD对TEC原始序列进行自适应分解,得到具有不同频率的分量并依据分量复杂度分析结果进行重构;其次,使用广义回归神经网络(GRNN)模型对高频分量进行建模与预报,使用Holt-Winters模型对低频分量进行建模与预报;最后,重构高频分量预报结果与低频分量预报结果得到电离层TEC预报值。根据太阳活动选取两段不同年积日、不同纬度电离层TEC序列进行实验,结果表明本文提出组合预报模型较单一的Holt-Winters模型、GRNN模型预报精度更高,在太阳活动平静期预报结果的平均相对精度为92.83%,在太阳活动剧烈期预报结果的平均相对精度为84.35%,对于长时间TEC预报也具有较好的效果,稳定性高。 In view of the shortcomings of the single ionospheric total electron content(TEC)prediction model,such as being easily interfered by external factors,and the prediction accuracy decreasing significantly with the increase of prediction time,this paper proposed a composite ionospheric TEC prediction model based on the complementary ensemble empirical mode decomposition(CEEMD).The key ways to achieve ionospheric TEC prediction by this model were:first,the original TEC sequence was adaptively decomposed by CEEMD to obtain components with different frequencies and then reconstructed according to the component complexity analysis results;Secondly,the general regression neural network(GRNN)model was applied to model and forecast high-frequency components,and the Holt Winters model was explored to model and forecast low-frequency components;Finally,the ionospheric TEC prediction values were obtained by reconstructing the high-frequency component prediction results and the low-frequency component prediction results.According to the solar activity,two periods of ionospheric TEC series with different annual product days and different latitudes were selected for experiments.The results showed that the prediction accuracy of the combined prediction model proposed in this paper was higher than that of the single Holt Winters model and GRNN model.The average relative accuracy of the prediction results in the quiet period of solar activity was 92.83%,and the average relative accuracy of the prediction results in the severe period of solar activity was 84.35%.It also had a good effect on long-term TEC prediction,High stability.
作者 金加棋 沈梁涛 JIN Jiaqi;SHEN Liangtao(Zhejiang Zhongce Xintu Geographic Information Technology Company Limited,Huzhou Zhejiang 313200,China)
出处 《北京测绘》 2023年第3期420-427,共8页 Beijing Surveying and Mapping
关键词 电离层 总电子含量 补充集合经验模态分解 Holt-Winters模型 广义回归神经网络模型 the ionosphere total electron content complementary ensemble empirical mode decomposition Holt-Winters model general regression neural network model
  • 相关文献

参考文献16

二级参考文献132

共引文献106

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部