Study on Exhaust Gas Temperature Prediction Model Based on GBDT

The real-time prediction of exhaust gas temperature is of great significance to the condition monitoring of aero engine. In this paper, a real-time EGT prediction model based on GBDT is proposed to solve the problem that the traditional machine learning model cannot provide sufficiently accurate prediction methods of EGT. Firstly, the idea of ensemble learning and the basic principle of GBDT algorithm are introduced. Secondly, combined with the demand of EGT prediction, the flow of the EGT prediction model based on GBDT is given. Finally, by using the actual QAR data of A320, a case study is carried out. The results of case study show that the model in this paper can predict EGT with a higher accuracy.