Chatter detection based on synchrosqueezing transform and statistical indicators in milling process

Chatter is a self-excited vibration between the workpiece and tool. In view of the non-stationarity of the chatter signal, the synchrosqueezing transform (SST) is used to process vibration signals during cutting, which can enhance the energy ratio of chatter. In order to eliminate the interference of tooth passing frequency and its harmonics, a time-frequency filtering method is applied to filter these frequency components out. Then, the vibration signal is reconstructed by inverse SST and statistical indexes in time and frequency domains are calculated. The cutting tests are carried out to select statistical indexes which are sensitive to chatter. The effectiveness of the proposed method is verified with cutting tests, and the results show that the chatter can be detected successfully before severe chatter marks are left on the workpiece.