Study on fault forecasting based on auto-regression model

As the demand of the modern equipment of prognostic and health management, fault forecasting is becoming more and more important. This paper study the fault forecasting method based on date, preprocess this data to meet the needs of the fault forecasting model, Then the algorithm based on auto-regression model are studied to predict fault. The experimental results show that the method is feasible. It can predict failure before equipment failure to provide the basis for on-condition maintenance system.