Tool Wear and Surface Quality Monitoring Using High Frequency CNC Machine Tool Current Signature

Benjamin Neef, Jonathan Bartels, Sebastian Thiede

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

7 Citations (Scopus)

Abstract

In this paper a machine learning approach for tool wear monitoring (TWM) and surface quality detection is proposed using high frequency current samples of a CNC turning machine main terminal. Significant frequency based features related to tool wear and surface quality are selected by univariate filter methods. Supervised machine learning methods including Support Vector Machine (SVM) and Random Forest Ensemble (RFE) are used to estimate tool wear and surface quality. Best hyper-parameter combinations of the proposed models are evaluated and found by grid search methods. Experimental studies are conducted on a CNC turning machine using a test work piece and the classification and accuracy results are presented. The presented methodology makes the set up of an on-line system for tool condition monitoring and an estimation of the work piece surface quality by the use of inexpensive and easy to install measurement hardware possible.

Original languageEnglish
Title of host publicationProceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018
PublisherIEEE
Pages1045-1050
Number of pages6
ISBN (Electronic)9781538648292
DOIs
Publication statusPublished - 24 Sep 2018
Externally publishedYes
Event16th IEEE International Conference on Industrial Informatics, INDIN 2018 - Porto, Portugal
Duration: 18 Jul 201820 Jul 2018
Conference number: 16

Conference

Conference16th IEEE International Conference on Industrial Informatics, INDIN 2018
Abbreviated titleINDIN 2018
CountryPortugal
CityPorto
Period18/07/1820/07/18

Keywords

  • Current
  • Electricity
  • Feature extraction and selection
  • Frequency domain
  • Main terminal
  • Tool wear monitoring
  • Turning

Fingerprint Dive into the research topics of 'Tool Wear and Surface Quality Monitoring Using High Frequency CNC Machine Tool Current Signature'. Together they form a unique fingerprint.

Cite this