基于BP神经络的极预报（.doc

基于BP神经网络的极移预报*收稿日期:2014-09-29
基金项目:航天科技创新基金(CASC201101),上海航天科技创新基金(SAST201251)
作者简介:张志(1983-),男,重庆忠县人,博士研究生,E-mail:neileva@;
廖瑛(通信作者),女,教授,博士,博士生导师,E-mail:liaoying1104@
张志1,廖瑛1,余越2
(,湖南长沙 410073; 4系,安徽蚌埠 233012)
摘要:为了提高地球定向参数极移的预报精度,建立了一个极移数据预报模型。利用傅里叶分析研究插值基础序列的周期特性,验证了基础序列重采样的可行性,提取插值基础序列数据的趋势项,利用多输入-单输出BP神经网络建模预报不同跨度的残差序列,合并趋势项和残差序列得到最终的极移预报。预报结果表明,选取合适的插值基础序列得到的预报极移精度较高,此BP神经网络能够有效地应用于地球定向参数极移的预报。
关键词:极移;傅里叶分析;BP神经网络;插值基础序列;趋势项;
中图分类号:P228 文献标志码:A 文章编号:
Prediction of polar motion based on BP work model
ZHANG Zhi1, LIAO Ying1, YU Yue2
( of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073, China;. 4 Department, Naval Petty Officer Academy, Bengbu 233012, China)
Abstract: A predictive model was set up to improve the prediction precision of polar motion of earth orientation parameters. The periodic characteristics of interpolated basic series was studied by fourier analysis, the feasibility of basic series resampling was verified, then the trend terms was derived from the interpolation basic series, and the multiple input single output BP work model was used to predict the residual series for different days into future, finally the predicted polar motion was achieved bining the trend terms with residual series. Prediction results indicate that the appropriate selection of int