A grinding chatter identification method based on a current instantaneous non-linear index

Abstract

The invention relates to a grinding chatter identification method based on a current instantaneous non-linear index. The method comprises the steps of 1) testing gear grinding machine grinding wheel drive motor currents synchronously and recording three phase alternating current signals; 2) converting the three phase alternating current signals into equivalent direct current signals and performing mean removing treatment; 3) performing continuous wavelet transform on the equivalent direct current signals subjected to the mean removing treatment; 4) calculating instantaneous wavelet bi-spectrums under different scales; 5) calculating phase randomness instantaneous wavelet bi-spectrums and at the same time performing repeated calculation by changing phase randomness parameters; 6) performing expectation taking operation on the instantaneous wavelet bi-spectrums which are calculated repeatedly for multiple times to obtain a phase randomness instantaneous wavelet bi-spectrum mean; 7) acquiring the phase randomness instantaneous bicoherence; 8) calculating the SNsWB index according to the phase randomness instantaneous wavelet bicoherence; 9) identifying grinding chatter according to the SNsWB index. Compared with the prior art, the method is high in accuracy and overcomes the defects of conventional methods in non-stable signal chatter identification method.

Description

technical field [0001] The invention belongs to the mechanical signal processing technical field of mechanical system health monitoring and control, and in particular relates to a grinding chatter identification method based on a current instantaneous nonlinear index. Background technique [0002] In the precision grinding process, since the depth of cut is usually very small, a small relative vibration between the workpiece and the grinding wheel will affect the grinding quality and accelerate the wear of the grinding wheel. Grinding chatter is a common problem in precision machine tool processing, which is mainly caused by mechanical self-excited vibration, and is also one of the main factors leading to unstable grinding. In recent years, scholars at home and abroad have carried out in-depth research on chatter recognition technology. Most of these studies are based on cutting force, vibration and noise information during machine tool processing. The methods involve Fourie...

AI Technical Summary

Problems solved by technology

However, the existing wavelet bicoherence model ignores the time information of the signal in order to improve the statistical stability and time resolution of the method, so it is not suitable for the analysis of non-stationary transient flutter signals

Images