Aero engine gas path fault prediction based on multi-sensor information fusion

For the problem of aero engine gas path fault diagnosis, the diagnosis results of RBF neural network, BP neural network and support vector machine (SVM) are fused at decision level with the D-S evidence theory, the results show that D-S evidence theory can achieve better diagnosis efficiency than the other three theories in separation, and it can reduce the misdiagnosis rate and improve the diagnostic performance. The fault prediction method based on information fusion can avoid the disadvantage of a single method. This method provides a determination to improve the reliability of aero engine, and the best maintenance decision reference and prolongs service life.