基于ARIMA模型的载人航天器氧分压分析及预测方法

Analysis and Predication of Manned Spacecraft Partial Oxygen Pressure Based on ARIMA Model

载人航天器环控生保系统中的氧分压分析监测关系到航天员在轨的安全与健康,是地面控制中心重点关注的关键信息。提出了一种基于时间序列数据模型的载人航天器氧分压分析及预测方法,对在轨遥测数据进行处理分析,应用ARIMA模型对氧分压历史数据进行分解建模,并预测其未来趋势。通过对载人航天器氧分压在轨数据的实测分析,对历史数据拟合均方根误差为01537 kPa,预测均方根误差为01378 kPa,预测精度较高。该方法基于短期历史环境信息分析建模,实现对未来状态变化的有效预测,有效提升了现有预测方法的预测时长,提前识别系统运行过程中的异常状态。

The analysis and monitoring of the partial oxygen pressure in ECLSS is related to the safe-ty and health of astronauts on orbit,and it is the key information that the ground control center needs to focus on.In this paper,a method for analyzing and predicting the partial oxygen pressure of man-ned spacecraft was proposed based on time series data model.The on-orbit environmental telemetry data were processed and analyzed.The ARIMA(Autoregressive Integrated Moving Average)model was used to model partial oxygen pressure.Data preprocessing,decomposition and modeling,fitting and prediction were used to analyze and model the historical data as well as predict the future trend of the environmental information.The results show that the root mean square error of fitting historical data is only 01537 kPa.The root mean square error of prediction is only 01378 kPa,which means it has high prediction accuracy.The method is based on the analysis and modeling of short-term his-tory information,and realizes effective prediction of state changes in the future,which improves the prediction time of the existing prediction methods.It is beneficial to identify the abnormal state in the process of system operation in.