Feature Extraction Method for Milling Cutter Wear Based on Optimized VMD

In view of the nonlinear and non-stationary features of vibration signals of milling processing as well as the features of being covered up by strong background noise, this paper proposes a feature extraction method for milling cutter wear based on optimized Variational Mode Decomposition (VMD). Since the decomposition effect of VMD is subject to the selection of penalty factor a and the number of decomposition modal component K, this research takes the minimization of Envelope Entropy as the indicator and adopts Differential Evolution (DE) for parameter adaptive optimization, which effectively solves the problem that the decomposition effect of VMD is subject to the selection of preset parameters. It is also more accurate and reliable than the subjective decision. According to the experimental results, the optimized VMD method, which can extract milling cutter wear features effectively, has a good noise reduction effect and a certain application value.