A Study On The Algorithm Of Signal Determination For Machined Surface Topography

The accuracy of the machine tool is the most important factor in determining the quality of the workpiece.In the machining process of the machine tool,various static and dynamic errors act on the workpiece forming process and are reflected to the workpiece surface in different degrees,which affects the surface quality.Therefore,noise reduction must be performed on the surface signal of the machine tool.However,traditional noise reduction methods can only perform local frequency signal processing,resulting in a large amount of signal information loss.Therefore,improving the decomposition accuracy of the signal components on the machined surface topography and improving the noise reduction effect are important topics for surface morphology research.This paper proposes an improved signal decomposition method based on the best wavelet packet basis for the first time.This method uses the information entropy value criterion to determine the optimal wavelet packet basis to make the energy distribution in the decomposed signal band more uniform,combined with the high-frequency harmonics addition method(HFHA-EMD)further improves the effect of Empirical Mode Decomposition(EMD)decomposition.The wavelet noise reduction method(WDCL)with compensation is proposed for the first time.This method introduces a compensation mechanism on the basis of traditional wavelet noise reduction,it reduces the loss of the real signal component and further removes noise by adjusting the noise reduction passing rate of different frequency components.Finally,the experimental verification of the improved signal decomposition method based on the best wavelet packet basis and the wavelet noise reduction method with compensation is carried out.The experimental results show that the new method can effectively improve the signal decomposition accuracy and the noise reduction effect.