Research on Performance Analysis of Aero-engine Based on QAR Time Series Data

Abstract

Flight record data is a typical type of time series data. Using data mining technology and artificial intelligence technology to analyze flight record data can assist relevant practitioners in timely maintenance. In flight record data, the trend of exhaust gas temperature (EGT) is an important parameter reflecting the change of aero-engine performance. Prediction of exhaust gas temperature by building an algorithmic prediction model, and then the difference between the predicted value and the measured value is analyzed to reflect the degree of aero-engine performance declined. This paper performs feature extraction and data cleaning on the flight data, and then establishes a prediction model. Finally, the LSTM algorithm is used to predict the multi-stage exhaust gas temperature, and compare with the actual data to calculate the root mean square error. The results show that the root mean square error increases from time, reflecting the trend of aero-engine recession. Provide a reference basis of maintenance as appropriate.