Automatic tool state identification in a metal turning operation using MLP neural networks and multivariate process parameters

On-Line and Indirect Tool Wear Monitoring in Turning With Artificial Neural Networks: A Review of More Than a Decade of Research

“…[84,146]). However, the general conclusion drawn in [55], 'Essentially, the fusion of different source signals of probable mediocre quality, yields better results than when only one such signal is used. ', is actually not correct.…”

“…Some authors demand the following minimum number of training patterns [41]: 'N ¼ 2ðn þ 1Þ where N is the number of training samples and n the number of input features.' The same heuristic is mentioned in [54,55,115,148,151], for instance. M .…”

In-process tool condition monitoring systems in CNC turning operations

“…However, to utilize the decomposed signals in a tool wear monitoring system, a decision-making system is necessary that classifies the amount of tool wear based on the information from the signals. Artificial Neural Networks (ANN) has been proposed as a decision-making system in tool condition monitoring systems (Chen & Jen, 2000;Choudhury, Jain, & Rao, 1999;Dimla Jr., Lister, & Leighton, 1998;Liu & Altintas, 1999;Niu, Wong, Hong, & Liu, 1998;Ôzel & Nadgir, 2002;Purushothaman & Srinivasa, 1994;Scheffer & Heyns, 2001;Silva, Reuben, & Wilcox, 1998;Tansel, Mekdeci, & McLaughlin, 1995;Venkatesh, Zhou, & Caudill, 1997;Zawada-Tomkiewicz, 2001). In the next section, the theoretical background and properties of ANN s are reviewed.…”

“…The accelerometer mounting position in turning operations has been proposed by a number of studies (Dimla Jr., Lister, & Leighton, 1998;Huang et al, 1999;O'Donnell et al, 2001;Schefer & Heyns, 2001). In this study, the sensor was mounted under the shank, determined to be an efficient position to detect the vibration from the cutting tool.…”

“…The acoustic emission (AE) sensor is another sensing technology that has been used in a number of studies (Li, 2002;Liang & Domfeld, 1989), but it is limited in its application by its noise integrity (Dimla Sr., 2000). Some studies adopted multi-sensor techniques to improve tool condition monitoring systems (Dimla Jr., Lister, & Leighton, 1998;Kuo, 2000;O'Donnell et al, 2001;Quan, Zhou, & Luo, 1998;Scheffer & Heyns, 2001;Silva, Reuben, Baker, & Wilcox, 1998). By combining multiple sensing technologies, these studies sought to develop more robust in-process tool condition monitoring systems.…”