航天器数据预测中的因果关系调整技术应用研究

在航天领域，预测参数不但受其自身时序变化的影响，也普遍受到关联因素的交叉影响。因果因素的影响也需要在时序预测过程中被考虑，该类问题被列入因果调整技术当中。为了更全面地反映航天器参数变化的实际情况，提出了一种基于层次分析法（AHP）的因果调整方法。该方法提供了因果因素的判断调整方法，并将调整系数应用于时序模型预测的修正。最终，建立了一个因果与时序预测有机结合的预测体系。

Reserch on Causal Adjustment Technology in Applying to Spacecraft Data Prediction

In space domain, predictive parameter is affected not only by its time series, also interactively by associative factors at large. When causal factors are encountered in the prediction horizon, their impact will be accounted for by causal adjustment technology. In order to entirely pursue the trace of spacecraft parameters in nature, this paper presents a method based on the Analytic Hierarchy Process to adjust prediction results. The method provides a facility to make judgmental adjustments and apply the adjustments to modification of time series model prediction. Finally, the paper builds a prediction architecture that integrated causal adjustment into time series prediction.