Tool wear monitoring in milling of titanium alloy Ti–6Al–4 V under MQL conditions based on a new tool wear categorization method
- 106 Downloads
Tool wear monitoring is crucial during machining of difficult-to-cut materials to save cost and improve efficiency. In this paper, a tool wear–monitoring strategy was proposed for milling of titanium alloy Ti–6Al–4 V under inner minimum quantity lubrication (MQL) conditions. Unlike the usual categorization method, tool wear was categorized into four states based on tool wear mechanism, tool wear rate, and tool life. Thus, more detailed information of tool could be predicted for tool wear monitoring. Cutting forces and acoustic emission were measured online as raw datasets. Statistical features were extracted from time and frequency domain, and mutual information (MI) was used for feature selection. Then, linear discriminant analysis (LDA) was adopted for dimensionality reduction and finding the optimal datasets for training. At last, ν-Support vector machine (ν-SVM) was applied for training and prediction. The proposed strategy had a prediction accuracy of 98.9%, which could be considered as valid and useful for tool wear monitoring.
KeywordsTool wear categorization Cutting forces Acoustic emission MQL Tool wear monitoring
The work is supported by the National Natural Science Foundation of China (No.51875355, 51675204 and 51875356), Shanghai Science and Technology Committee Major Program (17DZ1101202), and State Key Laboratory of Mechanical System and Vibration (MSVZD201801).
- 1.Davim JP (2014) Machining of titanium alloys. Springer, BerlinGoogle Scholar
- 25.Griffin J, Diaz F, Geerling E, Clasing M, Ponce V, Taylor C, Turner S, Michael E, Patricio F, Bronfman L (2017) Control of deviations and prediction of surface roughness from micro machining of THz waveguides using acoustic emission signals. Mech Syst Signal Process 85:1020–1034CrossRefGoogle Scholar
- 28.Qin W., Zha D., Zhang J. (2018) An effective approach for causal variables analysis in diesel engine production by using mutual information and network deconvolution. J Intell Manuf 1–11. https://doi.org/10.1007/s10845-018-1397-8
- 34.Atlas L, Ostendorf M, Bernard G (2000) In Hidden Markov models for monitoring machining tool-wear, In Proceedings of 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Istanbul. pp. 3887–3890Google Scholar