Acoustic Emission-Based Grinding Wheel Condition Monitoring Using Decision Tree Machine Learning Classifiers
Condition monitoring has emerged as an important technique in manufacturing industries for predictive maintenance and on-line monitoring of the processes and equipments. Due to the availability of sensors and signal processing technology, implementing condition monitoring systems in a manufacturing environment has become easy. In this paper, grinding wheel conditions in a surface grinding process are predicted with a simple decision tree-based machine learning classifier using time-domain acoustic emission signature. A grinding wheel attachment is designed and fabricated for capturing acoustic emission (AE) signal from the grinding wheel. Grinding wheel conditions are established using grinding wheel life cycle plot by monitoring surface roughness produced by the silicon carbide grinding wheel for the entire grinding cycle. AE signals were captured using the experimental set-up established for this study and statistical features are extracted from transients of AE. Classification and regression trees (CART) are used for establishing a correlation between AE features and grinding wheel conditions. The performance of the CART algorithms is evaluated using Gini index, towing and maximum deviation split criterions. Results indicate CART algorithms are efficiently predicting the grinding wheel condition with good accuracy.
KeywordsGrinding Condition monitoring Acoustic emission Decision tree
This research is supported by Directorate of Extramural Research and Intellectual Property Rights (ER & IPR), Defence Research and Development Organization (DRDO), ERIP/ER/0803740/M/01/1194, 13 January 2010.
- 7.Arun, A., Rameshkumar, K., Unnikrishnan, D., Sumesh, A.: Tool condition monitoring of cylindrical grinding process using acoustic emission sensor. Mater. Today: Proc. 5(5), 11888–11899 (2018)Google Scholar
- 10.Krishnakumar, P., Rameshkumar, K., Ramachandran, K.I.: Feature level fusion of vibration and acoustic emission signals in tool condition monitoring using machine learning classifiers. Int. J. Prog. Health Manag. 9, 1–15 (2018)Google Scholar
- 11.Krishnakumar, P., Rameshkumar, K., Ramachandran, K.I.: Machine learning based tool condition monitoring using acoustic and vibration data in high speed milling. Int. J. Intell. Decis. Technol. 1, 1–18 (2018)Google Scholar
- 12.Quinlan, J.R.: C4. 5: Programming for Machine Learning, vol. 38, p. 48. Morgan Kauffmann (1993)Google Scholar
- 13.Breiman, L.: Classification and Regression Trees. Routledge (2017)Google Scholar
- 14.Landis, J. R., Koch, G.G.: The measurement of observer agreement for categorical data. Biometrics, 159–174 (1977)Google Scholar