| 混沌特征参数对神经网络预测电离层TEC的影响分析 |
| |
| 引用本文： | 李淑慧,彭军还.混沌特征参数对神经网络预测电离层TEC的影响分析[J].测绘科学技术学报,2012(4):267-270. |
| |
| 作者姓名： | 李淑慧 彭军还 |
| |
| 作者单位： | 中国地质大学(北京)土地科学技术学院 |
| |
| 基金项目： | 国家自然科学基金项目(41104025;41074009);国土资源部公益性行业科研专项(200911015) |
| |
| 摘 要： | 利用国际GNSS服务组织(IGS)提供的东经115°经线上不同纬度处一年的电离层总电子含量(TEC)时间序列数据,研究了如何进一步提高基于神经网络方法预测电离层TEC的效果。研究表明:电离层TEC的预测误差与电离层TEC时间序列的最大Lyapunov指数与该序列均值的乘积具有较强的相关性;而且与时间延迟和嵌入维数的选择是否恰当也有着密切关系。
|
| 关 键 词： | 电离层 总电子含量 混沌 神经网络 预测 |
Impact Analysis of Chaotic Character Parameters on Neural Network Based Ionospheric TEC Prediction |
| |
| LI Shuhui,PENG Junhuan.Impact Analysis of Chaotic Character Parameters on Neural Network Based Ionospheric TEC Prediction[J].Journal of Zhengzhou Institute of Surveying and Mapping,2012(4):267-270. |
| |
| Authors: | LI Shuhui PENG Junhuan |
| |
| Institution: | (School of Land Science and Technology,China University of Geosciences(Beijing), Beijing 100083,China) |
| |
| Abstract: | Using the IGS(International GNSS Service) provided one-year time series of ionospheric TEC(Total Electron Content) data at different latitudes with the same longitude of 115° east,how to improve the predicted result of ionospheric TEC through Neural Networks methods was studied.Results showed that the prediction error greatly correlates with the product of the maximum Lyapunov exponent and the average value of TEC time series,and also have close relationship with the selection of the time delay and embedded dimension number. |
| |
| Keywords: | ionosphere TEC(Total Electron Content) chaos neural network prediction |
| 本文献已被 CNKI 等数据库收录！ |
