Needling machine transmission mechanism vibration signal feature extraction method

Abstract

The invention provides a needling machine transmission mechanism vibration signal processing method based on combination of MEEMD and an exponential wavelet threshold. The needling machine transmission mechanism vibration signal feature extraction method comprises the steps of: carrying out endpoint continuation processing on an acquired needling machine transmission mechanism vibration signal byadopting a waveform matched self-adaptive endpoint continuation method; performing wavelet decomposition, and denoising and reconstructing the acquired signal by utilizing an exponential wavelet threshold function to obtain the denoised signal; performing envelope line fitting on the denoised vibration signal by adopting a cubic B-spline method, and then carrying out MEEMD decomposition on the signal to obtain cross correlation coefficients and kurtosis values of a plurality of IMF components and the denoised signal; performing correlation analysis on the IMFs and the original signal to selectthe useful IMF; and finally performing Hilbert transform on the IMF components to obtain a Hilbert spectrum of the signal, and regarding the Hilbert spectrum as a signal feature of the needling machine transmission mechanism vibration signal. The needling machine transmission mechanism vibration signal feature extraction method is suitable for fault diagnosis of a needling machine transmission mechanism in the industrial field.

Description

technical field [0001] This application proposes a method for processing fault characteristic signals of transmission mechanisms of acupuncture machines based on the combination of MEEMD and exponential wavelet thresholds, which is suitable for fault diagnosis of transmission mechanisms of acupuncture machines in the mechanical field. Background technique [0002] With the continuous application of new technologies and materials, various technical performances of domestic needle looms have been greatly improved, but the basic principle is the same, that is, the needle plate equipped with needles is driven by the eccentric wheel linkage mechanism to operate. Reciprocate up and down to achieve the purpose of reinforcing the processed fiber web. The difference lies in the reciprocating positioning mechanism of the needle plate and the needle plate beam. In the design of domestic high-frequency acupuncture machines, there are mainly two types of mechanisms: one is the spindle bo...

AI Technical Summary

Problems solved by technology

[0007] The exponential wavelet threshold function processes the wavelet coefficients smaller than the threshold and the wavelet coefficients larger than the threshold, which solves the problem that the hard threshold function is discontinuous and has discontinuous points in the entire wavelet domain, reduces the variance of the noise reduction signal, and solves the problem of the soft threshold The problem with constant bias in functions

The combination of MEEMD and exponential wavelet threshold not only solves the problem of mode confusion in EMD decomposition, but also overcomes the problems of large amount of calculation, mode splitting, and white noise residue in EEMD decomposition and CEEMD decomposition.

Images