Rooker, T., Dervilis, N. orcid.org/0000-0002-5712-7323, Stammers, J. et al. (5 more authors) (2019) Predicting geometric tolerance thresholds in a five-axis machining centre. In: Niezrecki, C and Baqersad, J, (eds.) Structural Health Monitoring, Photogrammetry & DIC, Volume 6. 36th IMAC, A Conference and Exposition on Structural Dynamics 2018, 12-15 Feb 2018, Orlando, Florida. Conference Proceedings of the Society for Experimental Mechanics Series . Springer Nature , pp. 93-100. ISBN 978-3-319-74475-9
Abstract
NC-Checker is a software tool used for monitoring and validating the geometric performance in modern machining centres. Threshold settings allow the Manufacturing or Maintenance Engineer to customise the tool based on specific job or industry tolerance requirements. In order to perform effective long-term monitoring, this has the potential to skew the perceived health state of the machining centre as presented in the NC-Checker benchmark reports. This study brings attention to this fact and its relevance in the pursuit of enhanced levels of automation for geometric performance monitoring tools, in preparation for the machine shop’s transition to Industry 4.0. A sense-check function is proposed to identify unusual alterations based on historical data, utilising a support vector machine methodology to develop a predictive classifier. The models achieved predictive accuracy scores of 87.5% during validation, acquisition of a suitable testing set is under way and the predictive models will be evaluated upon completion.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Editors: |
|
Copyright, Publisher and Additional Information: | © The Society for Experimental Mechanics, Inc. 2019. This is an author-produced version of a paper subsequently published in Conference Proceedings of the Society for Experimental Mechanics. Uploaded in accordance with the publisher's self-archiving policy. |
Keywords: | Support vector machine; In-process inspection; CNC machining; Geometric performance; Condition monitoring |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Mechanical Engineering (Sheffield) |
Funding Information: | Funder Grant number Engineering and Physical Science Research Council (EPSRC) EP/R003645/1 |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 15 Jul 2019 11:08 |
Last Modified: | 16 Jul 2019 10:53 |
Status: | Published |
Publisher: | Springer Nature |
Series Name: | Conference Proceedings of the Society for Experimental Mechanics Series |
Refereed: | Yes |
Identification Number: | 10.1007/978-3-319-74476-6_14 |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:147483 |