Abstract
A pattern recognition strategy for on-line chatter detection was proposed and developed by analysing the acceleration signals from a milling operation. The components of the feature vectors were defined in terms of total power and dispersion of the dominant forced and chatter vibration modes extracted from linear stochastic models characterising the vibrations. The weighting vectors of a linear classifier were in turn evaluated by using the error correction method. Through the implementation of the method a significant rate of success could be demonstrated.
Original language | English (US) |
---|---|
Pages (from-to) | 279-290 |
Number of pages | 12 |
Journal | Mechanical Systems and Signal Processing |
Volume | 2 |
Issue number | 3 |
DOIs | |
State | Published - Jul 1988 |
ASJC Scopus subject areas
- Control and Systems Engineering
- Signal Processing
- Civil and Structural Engineering
- Aerospace Engineering
- Mechanical Engineering
- Computer Science Applications