A Novel Performance Prediction Method for Gas Turbines Using the Prophet Model

The performance prediction of gas turbines is one of the key factors to improve flight safety and reduce unnecessary maintenance costs. Although many performance prediction methods have already been developed, the prediction accuracy fails to meet the needs of practical application owing to several seasonal effects and change points in time series. To address these issues, this paper proposes a performance prediction method for gas turbines using the Prophet model. On the one hand, the model introduces the varying degradation rate into the trend item, which can better fit change points. On the other hand, the model can fit the seasonal effects using the Fourier series. Moreover, real monitoring data are used to verify the effectiveness of the proposed model in this paper. Experimental results show that compared with the ARIMA model and the SVR model, the proposed model achieves higher prediction accuracy, which can provide technical support for the maintenance plan formulation.