基于ARIMA和XGBoost算法的辅逆系统故障预测

Research on Fault Prediction of Auxiliary Inverse System based on ARIMA and XGboost Algorithm

提出XGBoost拟合各工况下辅逆温度变化曲线,利用拟合曲线与实际值残差做正态分布,确定警告阈值线和错误阈值线,用ARIMA算法预测后三天的趋势走向,并通过与警告阈值线和错误阈值线的比较给出相关预警的信息提示.

XGBoost is proposed to fit the auxiliary inverse temperature curve under various working conditions.A normal distribution is established by using the residual between the fitting curve and the actual value,and the warning and error warning lines are determined.The ARIMA algorithm is used to predict the trend in the next three days,and the warning and error warning lines are compared to give the warning information.