王喜江^1,,
边少锋¹,
李子申²,
蒋可¹,
任青阳³,
潘雄^1,,
1. 中国地质大学地理与信息工程学院, 武汉 430079
2. 中国科学院光电研究院, 北京 100094
3. 重庆交通大学土木工程学院, 重庆 400074

基金项目: 国家自然科学基金（41874009，41476087，41674043）联合资助


详细信息

作者简介: 王喜江, 男, 1995年生, 研究生, 主要从事电离层建模及预报方法研究.E-mail:wangxij95@163.com

通讯作者: 潘雄, 男, 1973年生, 教授, 主要从事现代测绘数据处理与应用的研究.E-mail:pxjlh@163.com

中图分类号: P352

收稿日期:2019-04-10
修回日期:2019-12-06
上线日期:2020-04-05

Prediction of global ionospheric TEC using the semiparametric kernel estimation method

WANG XiJiang^1,,
BIAN ShaoFeng¹,
LI ZiShen²,
Jiang Ke¹,
REN QingYang³,
PAN Xiong^1,,
1. School of Geography and Information Engineering, China University of Geosciences, Wuhan 430079, China
2. Academy of Opto-Electronics, Chinese Academy of Sciences, Beijing 100094, China
3. School of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China


More Information

Corresponding author: PAN Xiong,E-mail:pxjlh@163.com

MSC: P352

--> Received Date: 10 April 2019
Revised Date: 06 December 2019
Available Online: 05 April 2020


摘要
摘要:本文将半参数平差模型引入电离层球谐函数系数的预报中，建立了半参数球谐函数模型（Semiparametric-Spherical Harmonic，Semi-SH）来预测全球电离层总电子含量.首先，通过快速傅里叶变换获得球谐函数系数的周期和振幅，将振幅高的主周期归入趋势函数，振幅低的剩余周期归入随机信号，建立了半参数模型，同时利用核估计方法拟合趋势函数，解算随机信号，并在时间域上进行外推，得到了预报时间的球谐函数系数，代入15阶电离层球谐函数模型，最后得出电离层总电子含量（Total Electron Content，TEC）的预报值.本文基于欧洲定轨中心（CODE）发布的球谐函数系数进行电离层TEC长期预报和短期预报分析，其中长期预报采用四年预报两年的模式对球谐函数系数进行预报，短期预报设计了三个算例，采用前30天预报后一天的模式，分别预报1天、滑动预报7天和滑动预报30天.实验结果表明：长期预报能够较好地反映全球电离层TEC的变化趋势和波动情况，Semi-SH模型对全球电离层TEC平均值（Mean TEC global，MTEC_global）的拟合值和预报值与MTEC_global实际值的相关系数分别为0.8743和0.8010，呈现出高度相关性.短期预报中，在太阳活动高年和太阳活动低年，Semi-SH模型在中纬度地区预报精度较CODE发布的电离层TEC 1天预报产品（CODE'S 1-Day Predicted GIM，C1PG）有较大提升，在高纬度与低纬度地区两种模型预报精度相当；Semi-SH模型在太阳活动高年和太阳活动低年30天滑动预报精度的均值均高于C1PG模型.实验结果说明了Semi-SH模型预报电离层TEC值的有效性.
关键词: 电离层预报/
半参数球谐函数模型/
长期预报/
短期预报/
电离层总电子含量

Abstract:In this paper,the semiparametric adjustment model is introduced into the prediction of the ionospheric spherical harmonic coefficient,and a semiparametric-spherical harmonic(Semi-SH)model is proposed to predict the global ionospheric total electron content. Firstly,the period and amplitude of the spherical harmonic coefficient are obtained by fast Fourier transform,the main period with high amplitude is classified into the trend function,and the remaining period with low amplitude is classified into random signal,then establish the semiparametric adjustment model and use the kernel estimation method to fit the trend function,and calculate the random signal. The spherical harmonic coefficient of the predicted time was obtained by extrapolating in the time domain,and substitute into the 15-order ionospheric spherical harmonic function model. Finally,the predict value of total electron content (TEC) is obtained. In this paper,the long-term and short-term prediction analysis of ionospheric TEC is based on the spherical harmonic function coefficients published by the European Orbital Center (CODE). The long-term prediction uses the four-year forecast two-year mode to predict the spherical harmonic coefficient,and three examples are designed for short-term predict by using the mode of first 30 days predict the next day,the predict 1 day,sliding predict 7 days and sliding predict 30 days,respectively. The experimental results indicate that the long-term prediction can better reflect the trend and fluctuation of the global ionospheric TEC,the correlation coefficients of the fitted and predicted values of the global ionospheric mean TEC (MTEC_global) by Semi-SH model with the actual value of MTEC_global are 0.8743 and 0.8010,which are highly correlated. In the short-term prediction,the prediction accuracy of the Semi-SH model is greatly improved compared with the CODE'S 1-day predicted GIM (C1PG) released by CODE in the middle latitude region during the high solar activity and low solar activity. The prediction accuracy of the two models is quite in the high latitude and low latitude regions; the average of the sliding prediction accuracy of the Semi-SH model 30 days is higher than the C1PG model during the high solar activity and the low solar activity. The experimental results indicate the effectiveness of the Semi-SH model for predicting ionospheric TEC values.
Key words:ionosphere prediction/
semiparametric-spherical harmonic model/
long-term prediction/
short-term prediction/
ionospheric total electron content



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=dqwlxb_15405