Aero-engine rolling bearing fault diagnosis method based on residual network

The invention belongs to the field of aero-engine fault diagnosis, and particularly relates to an aero-engine rolling bearing fault diagnosis method based on a residual network. Comprising the following steps: S1, obtaining an aero-engine bearing fault test model, carrying out an aero-engine bearing fault test, and collecting a vibration acceleration signal; s2, constructing a residual network model; s3, marking the vibration acceleration signals according to different fault types; s4, performing Fourier transform on the marked vibration acceleration signal to obtain a spectrogram, and taking the spectrogram as the input of the residual network model; s5, dividing an original vibration acceleration signal into a training set and a test set, and training the residual network model; and S6, fault diagnosis of the engine rolling bearing is realized through the trained residual network model. The method does not need to depend on manual feature calculation, and has the characteristics of higher ...