Tool Wear State Recognition Based on Machine Learning

DOI: https://doi.org/10.62517/jike.202304209

Author(s)

Wenji Wang

Affiliation(s)

Zibo Vocational Institute, Zibo 255000,Shandong,China

Abstract

In order to quickly identify the current state of tool wear in real time, the standard of tool wear state division is analyzed first. The machine learning methods commonly used for pattern recognition classification including SVM, ANN, random forest and so on. In this paper, tool wear data sets are used to build a variety of machine learning models using the constructed feature space. LVQ, random forest and SVM are used to monitor the wear state respectively, and the classification accuracy of the classification model on the test set is calculated respectively.

Keywords

LVQ; Random Forest; SVM; Machine Learning State

References

[1] Tönshoff H K, Wulfsberg J P, Kals H J J, et al. Developments and trends in monitoring and control of machining processes[J]. CIRP Annals, 1988, 37(2): 611-622. [2] Rehorn A G, Jiang J, Orban P E. State-of-the-art methods and results in tool condition monitoring: a review[J]. The International Journal of Advanced Manufacturing Technology, 2005, 26(7-8): 693-710. [3] Tlusty J, Andrews G C. A critical review of sensors for unmanned machining[J]. CIRP annals, 1983, 32(2): 563-572.