Specific cutting energy: a physical measurement for representing tool wear
- 1 Downloads
In a machining context, unexpected tool breakage is still one of the primary causes of increase costs and machine downtimes. Hence, to increase productivity and ensure a company’s financial survival, a way to monitor cutting tool is essential. Thus, this paper proposes to show that it is possible to predict tool wear, with a low error, by using a recurrent neural network with a long short-term memory architecture. However, to achieve a general tool condition monitoring model, a proposition requiring no training is needed. Therefore, this paper introduces the concept of specific cutting energy based on the work of Debongnie , which is defined as the amount of energy required to remove 1 cm3 of material. Based on our work, we show that this feature is highly correlated (R > 90%) to the tool wear value. This concept also achieve a high adjusted R2 (R2 > 90%) with a linear regression model. These results are based on an experimental dataset provided by Agogino and Goebel . We succeed in achieving our objectives; however, future work should include a methodology to measure the residual useful life of a cutting tool based on the specific cutting energy and an industrial application of our methodology to see if the results support our conclusions. Still, our proposition could help machining companies accurately monitor their cutting tool wear condition with a single feature.
KeywordsTool condition monitoring Cutting energy EWMA chart Machine learning Recurrent neural network Long short-term memory architecture
Unable to display preview. Download preview PDF.
The authors would like to thank UC Berkely and NASA Ames Prognostic Data Repository for providing the Milling Dataset.
This work received financial support from the Fonds de Recherche du Québec – Nature et Technologies (FRQNT) through grant #257668.
- 1.Debongnie J-F (2006) Usinage. Editions du CEFAL,Google Scholar
- 2.Agogino A, Goebel K (2007) Milling Data Set https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository/. Accessed 23 May 2018
- 3.Kohler D, Weisz J-D (2016) Industrie 4.0 Les défis de la transformation numérique du modèle industriel allemand. FranceGoogle Scholar
- 4.Zhou Y, Xue W (2018) Review of tool condition monitoring methods in milling processes. Int J Adv Manuf Technol:1–15Google Scholar
- 5.Gebremariam MA, Xiang Yuan S, Azhari A, Lemma TA (2017) Remaining tool life prediction based on force sensors signal during end milling of Stavax ESR Steel. In: ASME 2017 International Mechanical Engineering Congress and Exposition, Tampa, Florida, USA, November 3-9, 2017. vol 58356. Volume 2: Advanced Manufacturing, p V002T002A094Google Scholar
- 8.Yu J, Liang S, Tang D, Liu H (2016) A weighted hidden Markov model approach for continuous-state tool wear monitoring and tool life prediction. Int J Adv Manuf Technol:1–11Google Scholar
- 9.Aghazadeh F, Tahan A, Thomas M (2018) Tool condition monitoring using spectral subtraction algorithm and artificial intelligence methods in milling process. IJMERR 7(1):30–34Google Scholar
- 10.Uekita M, Takaya Y (2017) Tool condition monitoring for form milling of large parts by combining spindle motor current and acoustic emission signals. Int J Adv Manuf Technol 89(1):65–75Google Scholar
- 12.Fu P, Hope A, King G (1998) Intelligent tool condition monitoring in milling operation.Google Scholar
- 15.Chen Y, Jin Y, Jiri G (2018) Predicting tool wear with multi-sensor data using deep belief networks. Int J Adv Manuf Technol:1–10Google Scholar
- 18.Goodfellow I, Bengio Y, Courville A (2016) Deep learning. The MIT Press,Google Scholar
- 19.Lei Y, Li N, Guo L, Li N, Yan T, Lin J (2018) Machinery health prognostics: a systematic review from data acquisition to RUL prediction. Mech Syst Signal Process 104:799–834Google Scholar
- 20.Thomas M (2011) Fiabilité, maintenance prédictive et vibration des machines. Presses de l’Université du Québec, MontréalGoogle Scholar
- 21.Aydin O, Guldamlasioglu S (2017) Using LSTM networks to predict engine condition on large scale data processing framework. In: 2017 4th International Conference on Electrical and Electronic Engineering (ICEEE), 8-10 April 2017. pp 281-285Google Scholar
- 22.Chollet F (2017) Keras (2015). https://keras.io. Accessed 28 Oct 2018