Time-frequency analysis method for time-varying multi-component signals of rotating machinery

The invention belongs to the field of signal processing, and particularly relates to a time-frequency analysis method for a time-varying multi-component signal of a rotating machine, which comprises the following steps: 1) acquiring the time-varying multi-component signal of the rotating machine to be analyzed; 2) processing an original signal based on a multi-scale wavelet threshold noise reduction algorithm of empirical mode decomposition; 3) carrying out initialization processing on the denoised signal based on short-time Fourier transform; 4) reconstructing an initial time-frequency distribution result through a time-frequency ridge line extraction method of EDSE, extracting a time-frequency ridge line based on a reconstructed time-frequency matrix and an extreme value search method, and obtaining an instantaneous frequency estimation value of the time-frequency ridge line; 5) separating a harmonic component from the denoised signal based on a Vold-Kalman filter and an instantaneous frequency estimation v