A Synchroextracting-Based Method for Early Chatter Identification of Robotic Drilling Process

Jianfeng Tao,Chengjin Qin,Chengliang Liu

DOI: https://doi.org/10.1007/s00170-018-2739-6

IF: 3.563

2019-01-01

The International Journal of Advanced Manufacturing Technology

Abstract:The robotic drilling process has shown its high industrial application potential for millions of holes in the aviation manufacturing. However, due to the relatively low stiffness of the serial robot manipulator, the robotic drilling system is more prone to chatter, leading to poor surface roughness, unbearable noise, and even causing severe damage to the end effector spindle. Consequently, it is of vital significance to identify and suppress this undesirable vibration. In this paper, a synchroextracting-based method is proposed for the early chatter detection of robotic drilling operations. The proposed algorithm is implemented through the following steps. First, the accurate time-frequency representation of the measured vibration signal is acquired by employing the synchroextracting transform (SET), which has high energy concentration and robustness to measurement noise. Second, the whole signal is divided into a finite number of frequency bands, and the corresponding sub-signal for each frequency band can be reconstructed by retaining the maximum coefficient of the SET. Finally, working as the chatter indicator, the statistical energy entropy is utilized to capture the inhomogeneous variation of energy distribution during the chatter transition process. Robotic drilling experiments with different cutting conditions were conducted to verify the effectiveness of the proposed chatter identification method. The results demonstrate that the presented synchroextracting-based algorithm can identify the chatter at an early stage, which is useful for the subsequent chatter suppression.