A novel denoising method for vibration signal of hob spindle based on EEMD and grey theory

Abstract

When the vibration analysis is used to evaluate the wear status of the hob cutter, the collection of vibration signals is inevitably contaminated by noise. To solve this problem, a novel denoising method for the vibration signal based on ensemble empirical mode decomposition (EEMD) and grey theory, named EEMD-Grey, is proposed in this paper. Grey relational analysis (GRA) and grey model (G-model), as two important techniques in grey theory, are adopted. In the EEMD-Grey method, the noisy signal is first decomposed into several intrinsic mode function (IMF) components by EEMD. Then, GRA is used to evaluate noise levels of IMF components. After that, noise-dominant IMF components are selected through G-model to remove noise. Finally, processed components and other components are reconstructed to obtain the denoised signal. Simulation experiments demonstrate the effectiveness of EEMD-Grey by comparison with other denoising methods. In addition, EEMD-Grey is applied to denoise the vibration signal of the hob spindle. The result shows that EEMD-Grey can effectively remove noise and retain useful information.