| Resumo: | The modern gas turbine engine widely used for aircraft propulsion is a complex integrated system that undergoes deterioration during operation due to the degradation of its gas path components. Turbofan engine-related costs can account for as much as a third of the total aircraft maintenance costs, which have driven the industry to adopt on-condition monitoring. In this work an integrated condition monitoring platform is proposed and developed for performance analysis and prediction. Different machine learning approaches are compared with the application of predicting engine behavior aiming at finding the optimal time for engine removal. The selected models were OLS, ARIMA, NeuralProphet, and Cond-LSTM. Long operating and maintenance history of two mature CF6 turbofan engines were used for the analysis, which allowed for the identification of the impact of different factors on engine performance. These factors were also considered when training the ML models, which resulted in models capable of performing prediction under specified operation and flight conditions. The ML models provided forecasting of the EGT parameter at the take-off phase allowing for predicting gradual performance deterioration under specified operation type. Cond-LSTM is shown to be a reliable tool for forecasting engine EGT with a MAE under 5℃. In addition, forecasting engine performance parameters has shown to be useful for identifying the optimal time for performing important maintenance action, such as engine gas path cleaning. This work has shown that engine removal forecast can be more precise by using sophisticated trend monitoring and advanced ML methods.
|
|---|