Classification of Vibration Data from Brownfield Milling Machines Using Machine Learning

Abstract

This study aims to classify vibration data obtained from old CNC milling (brownfield) machines used in industrial production processes with machine learning algorithms. The analysis of data obtained from such machines is of critical importance in order to increase the efficiency of old production machines and optimize production processes. In the study, vibration data collected from three different CNC machines under real production conditions for two years were used. The collected data were analyzed with various machine learning algorithms, especially overfitting prevention techniques, and the performances of these algorithms were compared. The results showed that the proposed machine learning methods can classify the information obtained from vibration data with high accuracy rates. The algorithms used provided an effective solution for early detection of tool wear, operational errors and other production problems caused by vibration, thus enabling more efficient management of production processes. The study presents an innovative method for modernizing old machines in particular within the framework of Industry 4.0, and provides important practical contributions in terms of improving industrial processes, optimizing maintenance processes and increasing overall efficiency.

Keywords

• CNC machines
• Vibration
• Machine learning
• Extreme learning
• Classification

References

1. Anonymous. 2020. Retrieved from bosch connected devices and solutions GmbH. URL: https://www.bosch-connectivity.com/%0Amedia/downloads/ciss/ciss_datasheet.pdf (accessed date: 23 September 2024).
2. Hesser DF, Markert B. 2019. Tool wear monitoring of a retrofitted CNC milling machine using artificial neural networks. Manuf Lett, 19: 1-4.
3. Huang G, Song S, Gupta JND, Wu C. 2014. Semi-supervised and unsupervised extreme learning machines. IEEE Trans Cybern, 44(12): 2405-2417.
4. Huang GB, Chen L. 2007. Convex incremental extreme learning machine. Neurocomputing, 70(16-18): 3056-3062.
5. Huang GB, Chen L, Siew, CK, 2006. Universal approximation using incremental constructive feedforward networks with random hidden nodes. IEEE Trans Neural Networks, 17(4): 879-892.
6. Hui Y, Mei X, Jiang G, Tao T, Pei C, Ma Z, 2019. Milling tool wear state recognition by vibration signal using a stacked generalization ensemble model. Shock Vib, 1: 7386523.
7. Lins RG, Guerreiro B, Schmitt R, Sun J, Corazzim M, Silva F. R, 2017. A novel methodology for retrofitting CNC machines based on the context of industry 4.0. 2017 IEEE Int Syst Eng Symp (ISSE), 1-6.
8. Lu Z, Wang M, Dai W. 2019. Machined surface quality monitoring using a wireless sensory tool holder in the machining process. Sensors, 19(8): 1847.

9. Nath C, 2020. Integrated tool condition monitoring systems and their applications: a comprehensive review. Procedia Manuf, 48: 852-863.
10. Quatrano A, De Simone MC, Rivera ZB, Guida D. 2017. Development and implementation of a control system for a retrofitted CNC machine by using Arduino. FME Trans, 45(4).
11. Tnani MA, Feil M, Diepold K. 2022. Smart data collection system for brownfield CNC milling machines: A new benchmark dataset for data-driven machine monitoring. Procedia CIRP, 107: 131-136.
12. Wszołek G, Czop P, Słoniewski J, Dogrusoz H. 2020. Vibration monitoring of CNC machinery using MEMS sensors. J Vibroeng, 22(3): 735-750.
13. Xiao D, Li B, Mao Y. 2017. A multiple hidden layers extreme learning machine method and its application. Math Probl Eng, 1: 4670187.
14. Zhang X, Dong S, Shen Q, Zhou J, Min J. 2023. Deep extreme learning machine with knowledge augmentation for EEG seizure signal recognition. Front Neuroinf, 17: 1205529.
15. Zhu W, Miao J, Qing L. 2014. Constrained extreme learning machine: a novel highly discriminative random feedforward neural network. Int Joint Conf Neural Networks (IJCNN), 800-807.
16. Zhu W, Miao J, Qing L, 2015. Constrained extreme learning machines: A study on classification cases. ArXiv Preprint ArXiv:1501.06115.