Aero-engine rolling bearing fault diagnosis method based on WPT-SVD

The invention provides an aero-engine rolling bearing fault diagnosis method based on WPT-SVD. The method comprises the steps that firstly, collected vibration signals are decomposed into different signal components, and the signal components are arranged into a matrix according to serial numbers; sVD decomposition is carried out on the matrix to extract singular values of all signal components, the singular values lower than the average level are zeroed, and inverse operation is carried out to restore a signal component matrix; dividing a high-energy signal component and a low-energy signal component according to the energy amplitude of the dimensionality reduction signal component, and screening a high-correlation signal component from the high-energy signal component and the low-energy signal component according to a Pearson correlation coefficient for superposition reconstruction; and carrying out fault diagnosis on de-noised signal feature extraction in an envelope demodulation mode. According to the met