Real-time chatter detection is crucial for the milling process to maintain the workpiece surface quality and minimize the generation of defective parts. In this study, we propose a new methodology based on the measurement of machine head stock structural vibration. A short-pass lifter was applied to the cepstrum to effectively remove components resulting from spindle rotations and to extract structural vibration modal components of the machine. The vibration modal components include information about the wave propagation from the cutter impact to the head stock. The force excitation from the interactions between the cutter and workpiece induces structural vibrations of the head stock. The vibration magnitude for the rigid body modes was smaller in the chatter state compared to that in the stable state. The opposite variation was observed for the bending modes. The liftered spectrum was used to obtain this dependence of vibration on the cutting states. The one-dimensional convolutional neural network extracted the required features from the liftered spectrum for pattern recognition. The classified features allowed demarcation between the stable and chatter states. The chatter detection efficiency was demonstrated by application to the machining process using different cutting parameters. The classification performance of the proposed method was verified with comparison between different classifiers.

In chatter detection, feature evaluation is an important task to identify mechanical systems and achieve higher classification accuracy. The importance of frequency bands is useful under various operating conditions. In this study, we propose a new methodology to identify the importance of frequency bands based on sub-band attention CNN. The sub-band attention CNN is a structure that combines the sub-band CNN and the attention layer. Unlike conventional CNNs that treat all frequency components with the same filter, the sub-band CNN processes different filters for each band. The attention layer is used to evaluate the importance of each band. The time-varying variance in frequency domain is used to extract chatter characteristics that vary greatly with time and it is used as an input for chatter detection. The useful frequency bands for chatter detection are obtained from the sub-band attention CNN. The importance of the frequency band is analyzed with the frequency response of the mechanical system.

No abstract