Tool Flank Wear Classification using Hilbert Huang Transform and Competitive Neural Network


Emerson Raja, J. and Venkataseshaiah, C. and Lim, W. S. (2012) Tool Flank Wear Classification using Hilbert Huang Transform and Competitive Neural Network. Australian Journal of Basic and Applied Sciences, 6 (10). pp. 61-71. ISSN 1991-8178

[img] Text
Restricted to Repository staff only

Download (299kB)


In this study, the relationship between emitted sound signal and the growth of tool wear was investigated and a new method is proposed for tool flank wear classification during turning operation. For this purpose, experiments were conducted in a turning machine in the university mechanical workshop by using fresh, slightly worn and severely worn carbide inserts while machining steel work piece. The emitted sound signal data was obtained by using a microphone. Tool wear was measured by a toolmaker’s microscope. The features namely, the instantaneous frequencies and their amplitudes, required for the competitive neural network to classify the state of the tool, were extracted from each emitted sound signal by using the new signal processing technique Hilbert Huang Transform. From the marginal spectrum plots, it is found that the increase in tool flank wear resulted in an increase of the sound pressure amplitude. This correlation enabled the competitive neural network to perform tool wear classification with 83.3% of accuracy.

Item Type: Article
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Faculty of Engineering and Technology (FET)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 19 Nov 2013 07:18
Last Modified: 19 Nov 2013 07:18


Downloads per month over past year

View ItemEdit (login required)