最优非负变权组合模型在卫星钟差预报中的应用

Application of the Optimal Non-Negative Variable Weight Combination Model for Satellite Clock Bias Prediction

针对单一卫星钟差预报模型自身存在的缺陷,提出基于二次多项式、灰色GM(1,1)和ARIMA等三种单一模型的最优非负变权组合预报模型。通过与各单一模型和经典权组合模型进行对比分析表明,该模型预报残差RMS值最小,精度更高,能够根据各模型预报效果赋予不同的权值,在一定程度上综合了多种单一模型的优点,可降低预报风险,提高模型预报的可靠性。

We consider the shortcomings of single satellite clock error prediction model. In this paper, based on three kinds of single model, i. e. , the quadratic polynomial, the gray GM （1, 1） and the ARIMA models, the optimal non-negative variable weight combination forecasting model is proposed. Comparison with the single model and the classic weight combination model, shows that the RMS val- ue of the optimal non-negative variable weight combination model prediction residual is smallest and its accuracy is higher. Moreover, it can give different weights according to the advantages and disad- vantages of each model and sum up the advantages of a variety of single models in a certain degree and reduce the forecast risk. Furthermore, it can also improve the reliability of the model forecast.