基于修正线性组合模型的原子钟钟差预报

Modified Linear Combination Model for Atomic Clock Prediction

原子钟钟差预报在时频工作中起着重要作用.目前常用的预报模型各有优缺点,为了综合各种预报模型特点,可以考虑组合这些模型预报的结果.针对线性组合模型预报钟差时存在的问题,提出了学习权的概念,以有效地利用各种精度信息,建立了修正组合预报模型.为了验证该方法的有效性,选取了4颗GPS卫星IGS(International GNSSService)钟差序列,组合了二次多项式模型和灰色模型的预报结果.结果表明,修正后的模型能在保证可靠性的基础上进一步提高结果的平稳性和准确性.

The atomic clock prediction plays community. Because the conventional prediction the results of these models can be combined to an important role in time and frequency models have advantages and limitations, synthesize the characteristics of various kinds of models. The combination model （CM） is constructed to predict the atomic clock. Actually, the results of CM are the weight averages of the results of single models. Con- sidering the problem occurring in long clock time series when using CM, learning weight is put forward to modify this model. Therefore, the weights of single models are relative to learning weight in the modified CM. To demonstrate the efficiency of this proposed method, the clock data of 4 GPS satellites are chosen and CM is used to combine the results from the quadratic polynomial model and grey model. The result shows that the reliability is improved when CM is adopted. In comparison with CM, the modified CM can remarkably improve the stability and precision of the atomic clock prediction.