基于时间序列的航天器遥测数据预测算法

Spacecraft Telemetry Data Prediction Algorithm Based on Time Series

摘要:

在航天器遥测数据预测领域,基于时间序列的预测方法有着广阔的应用前景;时间序列有一明显的特性就是记忆性,记忆性是指时间数列中的任一观测值的表现皆受到过去观测值影响;它的基本思想是根据观测数据的特点为数据建立尽可能合理的统计模型,利用模型的统计特性解释数据的统计规律,以期达到预报的目的;提出了采用模式识别和参数估计的方法,结合航天器遥测动态数据,建立关于航天器遥测数据的时序预测模型,对航天器遥测数据趋势进行检测和预报。

Abstract:

The prediction method based on time series has broad application prospects in the field of spacecraft telemetry data prediction. One obvious characteristic of time series is memory, and memory is the performance of any observed value in the time series. The basic idea of this model is to establish the statistical model of the data according to the characteristics of the observation data, and to use the statistical characteristics of the model to explain the statistics of the data, so as to achieve the purpose of forecasting. The algorithm using pattern recognition and parameter estimation method, and combined with the spacecraft telemetry dynamic data, the establishment of spacecraft telemetry data time series prediction model, to detect and forecast the trend of the spacecraft telemetry data.