摘要
针对目前极移最小二乘(Least Square,LS)+自回归(AutoRegressive,AR)预报模型的单一数据选取方案,提出分别考虑LS模型数据量和AR残差数据量的组合数据模式,并对极移预报时单一数据和组合数据预报结果精度进行分析,探讨模型输入数据量对极移预报精度的影响.结果表明,模型输入数据量的变化对极移预报结果影响较大.采用组合数据预报的方式相比较于单一数据量预报方式精度更高,特别是针对30--360 d跨度内的中长期预报,组合数据量的极移预报精度可比单一数据量预报精度有较大改善.结论证明组合数据在极移预报时具有一定的优势,可为以后极移预报数据量选取提供一定的借鉴参考意义.
Based on the single data input method of the current LS(Least Square)+AR(AutoRegressive)model of polar motion forecasting,this paper considers a combined data mode of the sequence length of the LS model and the AR residual data separately.The single data and the combined data are respectively used for forecasting,and then the forecast accuracy is analyzed,and the influence of the model input data volume on the accuracy of the polar motion forecast is discussed.The results show that the change in the amount of input data of the model has a greater impact on the prediction results of polar motion.The combined data forecasting method has higher accuracy than the single data,especially for the medium and long-term forecasts within a span of 30 to 360 days.The combined data forecast accuracy can be greatly improved compared to the single data forecasting accuracy.The conclusion proves that the combined data has certain advantages in the prediction of polar motion,and it can provide a certain reference for the selection of data volume of polar motion forecast in the future.
作者
徐海龙
乔书波
王穆阳
XU Hai-long;QIAO Shu-bo;WANG Mu-yang(Institute of Surveying and Mapping,Information Engineering University,Zhengzhou 450001;31121 Troop,Nanjing 210018)
出处
《天文学报》
CAS
CSCD
北大核心
2022年第2期11-18,共8页
Acta Astronomica Sinica
基金
国家自然科学基金项目(42074010、42104033)资助。
