GPS卫星钟差组合预报模型研究  

Research on GPS Satellite Clock Error Combined Prediction Model

作　　者：刘伟昌 周志广 LIU Weichang;ZHOU Zhiguang(Zhejiang Institute of Surveying and Mapping Science and Technology,Hangzhou 311100,China)

机构地区：[1]浙江省测绘科学技术研究院,浙江杭州311100

出　　处：《测绘与空间地理信息》2023年第10期118-120,124,127,共5页Geomatics & Spatial Information Technology

摘　　要：为了提高卫星钟差预报的稳定性与精度,本文结合幂函数变换GM(1.1)模型与BP(Back Propagation)神经网络模型在卫星钟差预报中的优势,提出了一种卫星钟差组合预报模型。该组合预报模型首先使用幂函数变换GM(1.1)模型对卫星钟差进行建模,获取预报残差值;其次,将幂函数变换GM(1.1)模型的预报残差值作为BP神经网络模型的输入值进行BP神经网络模型构建;最后,将两种模型预报结果进行重构得到最终预报结果。为了对本文提出的组合预报模型的有效性与优越性进行检验,使用IGS站提供的两颗卫星高精度钟差数据进行模型验证,使用均方根误差RMSE作为衡量模型预报性能的评价指标,结果表明,相比于幂函数变换GM(1.1)模型预报结果与BP神经网络模型预报结果,本文组合预报模型对PRN03、PRN08卫星钟差的均方根误差均有所降低,预报精度更高,表现出了更好的钟差预报性能,可为实际卫星钟差预报、提高单点定位精度提供借鉴与参考。In order to improve the stability and accuracy of satellite clock error prediction,combined with the advantages of power function transformation GM(1.1)model and BP(Back Propagation)neural network model in satellite clock error prediction,a satellite clock error combined prediction model is proposed in this paper.Firstly,the combined prediction model uses the power function transformation GM(1.1)model to model the satellite clock error and obtain the forecast residual.Secondly,the prediction residual value of power function transformation GM(1.1)model is used as the input value of BP neural network model to construct BP neural network model;Finally,the prediction results of the two models are reconstructed to obtain the final prediction results.In order to test the effectiveness and superiority of the combined prediction model proposed in this paper,the model is verified by using the high-precision clock error data of two satellites provided by IGS station,and the root mean square error RMSE is used as the evaluation index to measure the prediction performance of the model.The results show that compared with the prediction results of power function transformation GM(1.1)model and BP neural network model,in this paper,the combined prediction model reduces the root mean square error of PRN03 and PRN08 satellite clock error,has higher prediction accuracy,and shows better clock error prediction performance,which can provide reference for the actual satellite clock error prediction and improve the single point positioning accuracy.

关 键 词：幂函数变换GM(1.1)模型 BP神经网络模型 组合预报模型 卫星钟差 

分 类 号：P228[天文地球—大地测量学与测量工程]