Condition monitoring of cutting tools using artificial neural networks

Gindy, N. and Al-Habaibeh, A. ORCID: 0000-0002-9867-6011, 1997. Condition monitoring of cutting tools using artificial neural networks. In: A.K. Kochhar, ed., Proceedings of the Thirty-Second International Matador Conference. London: Palgrave, pp. 299-304. ISBN 9781349146222

Preview

Text 1640479_Al-Habaibeh.pdf - Published version Download (3MB) | Preview

Abstract

The paper presents a methodology for using neural network techniques and simple data processing algorithms for monitoring the condition of milling cutters during peripheral milling . The learning algorithms considered in this research utilise artificial neural networks to map some machining parameters to sensory signals. Cutting force and acceleration signals recorded during machining are first simplified and then fed into the input layer of the neural network. Using the back-propagation method, the output of the neural network is used to recognise “normal” as well as “faulty” milling cutters and the depth of cut used. The experimental results show that the proposed approach of using simple data processing algorithms with neural networks is capable of successfully identifying common fault conditions in milling cutters in peripheral operations.

Item Type:	Chapter in book
Description:	Paper presented at the Thirty-Second International Matador Conference, Manchester, 10-11 July 1997.
Creators:	Gindy, N. and Al-Habaibeh, A.
Publisher:	Palgrave
Place of Publication:	London
Date:	1997
ISBN:	9781349146222
Identifiers:	Number Type 10.1007/978-1-349-14620-8_47 DOI 1640479 Other
Rights:	© Department of Mechanical Engineering, University of Manchester Institute of Science and Technology, 1997.
Divisions:	Schools > School of Architecture, Design and the Built Environment
Record created by:	Linda Sullivan
Date Added:	09 Feb 2023 12:24
Last Modified:	09 Feb 2023 12:24
URI:	https://irep.ntu.ac.uk/id/eprint/48221

Actions (login required)

Edit View

Views

Downloads