Tool wear predictability estimation in milling based on multi-sensorial data

The safe and reliable operations in industrial manufacturing processes play a crucial role in the economic productivity. Machining process disturbances such as collision, overload, breakdown, and tool wear tend to cause production system failures. The current study aims at investigating the limitations of tool wear prediction on the milling of CGI 450 plates, through the simultaneous detection of acceleration and spindle drive current sensor signals. Tool wear prediction has been accomplished, by utilizing the experimental results that derived from third degree regression models and pattern recognition systems. These results indicate that predictability is affected by the mean signal energy, acquired from the vibration acceleration signals.