Analysis of the entropy randomness index for machining chatter detection

作者：Perez C****es Daniel; Vela Martinez Luciano; Carlos Jauregui Correa Juan; Alvarez Ramirez Jose^*

来源：International Journal of Machine Tools and Manufacture, 2012, 62: 39-45.

DOI：10.1016/j.ijmachtools.2012.06.007

• Chatter
• Entropy
• Randomness

This paper presents an approach for chatter monitoring based on a proper tuning of the approximate entropy (AE) method. The underlying idea behind the approach is that unstable chattering is related to the emergence of random dynamics, while stable chattering is dominated by regular periodic dynamics. In fact, unstable chatter is characterized by the complex interaction of stochastic subcritical Hopf and period doubling bifurcations, as well as limit cycles, quasiperiodic and even chaotic behavior. In this way, as the working condition of a machine system deteriorates due to changes in the operation conditions or even the progression of structural defects, many spectral components contained in the chatter signal will dominate the machine dynamics, resulting in increments of the randomness content of the acceleration signal. The ability of the approach for monitoring chatter evolution is verified experimentally through vibration measurements on a realistic milling process. The study shows that AE can effectively characterize the severity of chatter instabilities, with good computational efficiency and high robustness.

• 出版日期2012-11