Zobrazit v EDS

Building digital-twin virtual machining for milling chatter detection based on VMD, synchro-squeeze wavelet, and pre-trained network CNNs with vibration signals.

Uloženo v:
Podrobná bibliografie
Název: Building digital-twin virtual machining for milling chatter detection based on VMD, synchro-squeeze wavelet, and pre-trained network CNNs with vibration signals.
Autoři: Jauhari, Khairul, Rahman, Achmad Zaki, Al Huda, Mahfudz, Widodo, Achmad, Prahasto, Toni
Zdroj: Journal of Intelligent Manufacturing; Oct2024, Vol. 35 Issue 7, p3083-3114, 32p
Témata: CONVOLUTIONAL neural networks, ARTIFICIAL intelligence, COMPUTERS, VIRTUAL machine systems, SIGNAL processing
Abstrakt: Smart machining is becoming a major trend in the present manufacturing industry, which is increasingly adopting digital technology and artificial intelligence to improve production processes' quality, speed, efficiency, and safety. This condition leads to the presentation of an updated study regarding the application of the digital-twin virtual machining model development, to detect chatter phenomena in milling processes. Chatter is a dynamic interaction where an instability state is observed between the workpiece and the cutter during material removal. This process affects the roughness of the finish surface and tool-life and eventually reduces the machining results in quality. Consequently, a novel intelligent machining system was developed for detecting chatter based on the variational mode decomposition method, wavelet-based synchro-squeeze transform, and Transfer Learning (TL) application. This TL application was created using modified pre-trained convolution neural networks to identify unstable (chatter) or stable state conditions of the process of milling-cut. The model was also developed due to being data-driven, where the measured vibration signals for the process of milling-cut were trained and tested through several modified pre-trained networks. The results showed a good-level model with an average classification accuracy of 94.04%. Therefore, the manufacturing industry could adopt this novel method, especially in machining, to overcome the problem of emphasizing the limited process of data monitoring conditions. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Intelligent Manufacturing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Databáze: Complementary Index
Popis
Abstrakt:Smart machining is becoming a major trend in the present manufacturing industry, which is increasingly adopting digital technology and artificial intelligence to improve production processes' quality, speed, efficiency, and safety. This condition leads to the presentation of an updated study regarding the application of the digital-twin virtual machining model development, to detect chatter phenomena in milling processes. Chatter is a dynamic interaction where an instability state is observed between the workpiece and the cutter during material removal. This process affects the roughness of the finish surface and tool-life and eventually reduces the machining results in quality. Consequently, a novel intelligent machining system was developed for detecting chatter based on the variational mode decomposition method, wavelet-based synchro-squeeze transform, and Transfer Learning (TL) application. This TL application was created using modified pre-trained convolution neural networks to identify unstable (chatter) or stable state conditions of the process of milling-cut. The model was also developed due to being data-driven, where the measured vibration signals for the process of milling-cut were trained and tested through several modified pre-trained networks. The results showed a good-level model with an average classification accuracy of 94.04%. Therefore, the manufacturing industry could adopt this novel method, especially in machining, to overcome the problem of emphasizing the limited process of data monitoring conditions. [ABSTRACT FROM AUTHOR]
ISSN:09565515
DOI:10.1007/s10845-023-02195-0