CAJ | 학술논문

精密卫星钟差是精密定轨的重要因素之一．将BP神经网络模型引入卫星钟差预报，分别对少量及大量的卫星钟差数据进行预报，将其预报精度和灰色模型GM（1，1）预报结果进行分析比较，证明BP神经网络模型适于卫星钟差预报．通过对IGU超快速星历中的卫星钟差进行预报，发现前期预报精度很高，验证了其实际有效性．
정밀위성종차시정밀정궤적중요인소지일．장BP신경망락모형인입위성종차예보，분별대소량급대량적위성종차수거진행예보，장기예보정도화회색모형GM（1，1）예보결과진행분석비교，증명BP신경망락모형괄우위성종차예보．통과대IGU초쾌속성력중적위성종차진행예보，발현전기예보정도흔고，험증료기실제유효성．
The precise satellite clock bias is one of the most important fneural netwoactors for satellite orbit determination. In this paper, the BP rk model was used for satellite clock bias forecast, using a vast amount of data and a few data respectively. Then the forecast accuracy of BP model was compared with that of gray model GM (1, 1). The results suggest that the BP neural network model is suitable for the satellite clock bias forecast. Further more, the BP neural network model was used for forecast based on ultra-rapid clock bias data, and the high precision in prophase indicated it is effective.