灰预测是灰色系统理论的重要组成部分,已成功应用于若干领域的预测;时间序列分析是定量预测方法之一;研究了利用以灰预测模型为基础,建立的残差灰预测模型和AR时间序列模型对航天器故障状态进行预测的方法;首先分析了灰预测模型以及残差灰色预测模型的建立过程,之后分析了AR时间序列模型的建立过程;然后利用所建立的模型进行航天器故障状态的预测,给出预测结果;研究表明,残差灰预测模型预测误差比基本灰模型降低,在特定的数据特征条件下也要优于AR时间序列模型.%Gray prediction, one of the key parts of gray system theory, has been used in several fields. Time series analysis is one of the quantitative forecasting methods. Here, the methods to predict fault status of satellite based on gray model, residual gray model and AR time series model have been studied. Firstly, the gray model and the residual gray model are analyzed, and then the AR time series model is analyzed. Finally, use the models to predict fault status of spacecraft, and the prediction result is analyzed. It shows the residual gray model is superior to the basic gray model and the AR time series model.
展开▼