Order tracking method based on nonlinear frequency modulation wavelet transformation

Abstract

The invention relates to an order tracking method based on nonlinear frequency modulation wavelet transformation. The order tracking method comprises the steps of: firstly, acquiring a vibration signal of a measured rolling bearing through an acceleration or velocity sensor mounted thereon, adopting nonlinear frequency modulation wavelet transformation for carrying out time-frequency analysis on the vibration signal to obtain joint time-frequency distribution of the signal, searching a maximum spectral peak at each moment in the joint distribution to obtain a corresponding instantaneous frequency value, fitting the instantaneous frequency values by utilizing a instantaneous frequency value, and regarding an obtained value as an estimated value of instantaneous frequency; and determining a center and bandwidth of resonant frequency of the vibration signal by utilizing spectral kurtosis, adopting a band-pass filtering method to extract an inherent resonant component, carrying out envelope demodulation on the inherent resonant component, and finally performing uniform angle resampling on the envelope signal by utilizing instantaneous frequency to obtain an order spectrum of the rolling bearing and complete order tracking of the signal.

Description

technical field [0001] The invention belongs to the field of fault diagnosis and signal processing of rotating machinery, in particular to an order tracking method based on nonlinear frequency modulation wavelet transform. Background technique [0002] Due to the wide application of rolling bearings in rotating machinery, the running state directly affects the running accuracy, reliability and service life of the whole machine. Therefore, the fault diagnosis and prediction of rolling bearings is very important. Envelope analysis is a common method for extracting fault features of rolling bearings, and the variable-speed operating conditions of rolling bearings are very common. In this condition, the envelope of the vibration signal shows aperiodicity and non-stationarity. Using traditional spectrum analysis to process non-stationary signals often exhibits frequency ambiguity. [0003] The order tracking technology can avoid the aliasing of the separated frequency component...

AI Technical Summary

Problems solved by technology

Patent application numbers 201310256814.7 and 201510043916.X propose to use the short-time Fourier transform to analyze the time-frequency of the signal and then estimate its instantaneous frequency, but the short-time Fourier transform used in this method is subject to the principle of uncertainty. There is a problem of low time-frequency resolution, which leads to inaccurate instantaneous frequency estimation

Images