Data-Driven Exhaust Gas Temperature Baseline Predictions for Aeroengine Based on Machine Learning Algorithms

The exhaust gas temperature (EGT) baseline of an aeroengine is key to accurately analyzing engine health, formulating maintenance decisions and ensuring flight safety. However, due the complex performance characteristics constraints many external factors, it difficult obtain accurate non-linear features between various operating factors EGT. In order diagnose forecast quickly accurately, four data-driven prediction frameworks for EGT are proposed. These took conditions state control parameters as input variables predicted output variables. original data were collected from CFM56-5B ACARS data. Four typical machine learning methods, including Generalized Regression Neural Network (GRNN), Radial Basis (RBF), Support Vector (SVR) Random Forest (RF) trained develop models. models validated by comparing after-flight another engine. results show that developed GRNN have best accuracy computational efficiency compared with other models, their RE CPU calculation time on verification set 1.132 × 10?3 3.512 s, respectively. can meet needs practical engineering applications airlines. methodologies be employed airlines predict purpose monitoring health management.