Chatter identification in milling of the thin-walled part based on complexity index

In order to avoid chatter in milling of thin-walled part, chatter prediction based on the complexity of vibration signals in the milling process is proposed. According to the nonlinear characteristics of vibration signals in milling process, the complexity which is a nonlinear dynamics index and has higher computational efficiency is used to as the chatter prediction index in milling of thin-walled part, and the physical meaning of the complexity index is explained in detail. Firstly, different types of chatters in the milling process are simulated by numerical calculation, and the causes that cause different types of chatter are discussed. Secondly, according to the chatter types in the experiment of the paper, the complexity is used to analyze the vibration signals simulated by tending to the flip bifurcation, the result show that complexity index is valid in the chatter prediction for the simulated vibration signals by tending to chatter. In order to verify the validity of the complexity for the chatter prediction, the experimental platform in milling of the thin-walled part was set up. The complexity is used to analyze the actual vibration signals caused by tending to the chatter, and the results show that the index is effective in the actual milling chatter prediction.