Milling cutting tool diagnosis using comparisons of the excitation identified by cepstral techniques

Sghir, K. A., Randall, R. B., Badaoui, M. E. and Guillet, F. (2007). Milling cutting tool diagnosis using comparisons of the excitation identified by cepstral techniques. In: Martin Veidt, Faris Albermani, Bill Daniel, John Griffiths, Doug Hargreaves, Ross McAree, Paul Meehan and Andy Tan, Proceedings of the 5th Australasian Congress on Applied Mechanics (ACAM 2007). 5th Australasian Congress on Applied Mechanics (ACAM 2007), Brisbane, Australia, (737-742). 10-12 December, 2007.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
P2.1.pdf Session P2.1: cepstral techniques: Sghir papers application/pdf 324.73KB 801
Author Sghir, K. A.
Randall, R. B.
Badaoui, M. E.
Guillet, F.
Title of paper Milling cutting tool diagnosis using comparisons of the excitation identified by cepstral techniques
Conference name 5th Australasian Congress on Applied Mechanics (ACAM 2007)
Conference location Brisbane, Australia
Conference dates 10-12 December, 2007
Proceedings title Proceedings of the 5th Australasian Congress on Applied Mechanics (ACAM 2007)
Place of Publication Brisbane
Publisher Engineers Australia
Publication Year 2007
Year available 2008
Sub-type Fully published paper
ISBN 0 8582 5862 5
Editor Martin Veidt
Faris Albermani
Bill Daniel
John Griffiths
Doug Hargreaves
Ross McAree
Paul Meehan
Andy Tan
Volume 1
Start page 737
End page 742
Total pages 6
Collection year 2007
Language eng
Abstract/Summary This paper investigates the diagnosis of cutting tools in a milling operation using vibration signals and proposes a signal processing algorithm to achieve that. In the proposed algorithm, the impulse response of the measured vibration signal is firstly identified using the random decrement technique. This is then converted to a cepstrum and subtracted from the measured signal in the quefrency domain using the additive properties of cepstra. The residual signal representing the forcing function is then transformed back into the time domain using the inverse cepstrum. Finally the power spectral density is estimated, and a comparison is made between the different states of the cutting tool. For a good estimation of the force, four measurement points are used, and the identified excitation sources are then averaged. By comparing the spectra of the forcing functions, the efficiency of the method is demonstrated, and the faulty case is clearly distinguished from the fault-free case. This was not the case with the original response signals.
Subjects 290501 Mechanical Engineering
Keyword Random decrement
complex cepstrum
blind identification
diagnosis
milling cutting tools
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status Unknown

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Thu, 13 Mar 2008, 10:13:04 EST by Laura McTaggart on behalf of School of Engineering