A Burr Location and Elimination Method Based on Direct Method to Solve Instantaneous Frequency

Abstract

The invention discloses a burr location and elimination method based on resolution of instantaneous frequency. The method comprises the following steps: decomposing a signal into a product function component by a partial mean decomposition method to obtain a frequency modulation signal of a product function, setting all local maximum points of the frequency modulation signal set to 1, and settingall local minimum points to 1; performing inverse cosine on the frequency modulation signal to obtain a non-monotone instantaneous phase, expanding the non-monotone instantaneous phase to obtain a monotonously-increasing instantaneous phase; performing forward phase difference on the monotonically-increasing instantaneous phase to obtain burred instantaneous frequency of the frequency modulation signal; and determining the burr position of the burred instantaneous frequency; deleting data of the burred instantaneous frequency on a burr position sequence, and interpolating the instantaneous frequency value of the burr position sequence with y three splines in order to obtain instantaneous frequency without burr. The method has the advantages of no parameters no dependence on the user experience, and has the characteristic of local processing.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and in particular relates to a burr location and elimination method for solving instantaneous frequency based on a direct method. Background technique [0002] Instantaneous frequency is an important modulation parameter of amplitude modulation-frequency modulation (AM-FM) signals, which is of great significance for exploring the detailed mechanism of non-stationary and nonlinear processes. At present, the idea of ​​solving the instantaneous frequency is to first decompose the signal into a series of single-component signals, and then demodulate these single-component signals. The more commonly used ones include Hilbert-Huang transformation (NordenE.Huang.etc.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the royal society A:Mathematical,Physical and Engineering Sciences.1998: pp903-995.), empirical AM-F...

AI Technical Summary

Problems solved by technology

Because the burr often corresponds to the high frequency component in the IF curve, some scholars (Ren Daqian, Yang Shixi, Wu Zhaotong, etc. compared the signal instantaneous frequency direct measurement method with the Hilbert transform and Teager energy method[J]. Chinese Journal of Mechanical Engineering, 2013 ,49(9):42-48.) used the moving average method to smooth the obtained IF, but did not give the selection method of smoothing parameters including smoothing times and smoothing steps; some scholars also used the morphological filtering method (Tang Guiji , Wang Xiaolong. LabVIEW implementation of LMD and its application in gear fault diagnosis [J]. China Test, 2014.40(1): p101-105.) to eliminate the distortion in the instantaneous frequency obtained by the direct method, also not Parameters such as structural elements and operation types used are given, and these methods are all indiscriminate processing of the entire instantaneous frequency sequence. While processing the glitch position, it also affects the normal value of the non-glitch position, which has not been seen so far. A method that can accurately locate these burr points and eliminate them at fixed points

Images