No abstract

No abstract

The objective of this study is to establish a signal processing methodology that can infer the state of milling insert wear from translational vibration measured on the spindle housing of a milling machine. First, the tool wear signature in a translational vibration is accentuated by mapping the translational vibration into a torsional vibration using a previously identified nonlinear relationship between the two, i.e., a virtual sensor. Second, a time-frequency distribution, i.e., a Choi-Williams distribution, is calculated from the torsional vibration. Third, scattering matrices and orthogonalization are employed to identify the time-frequency components that are best correlated to the state of wear. Fourth, a neural network is trained to estimate the extent of wear from these critical time-frequency components. The combination of the virtual sensor, time frequency analysis and neural network is then validated with data obtained from real cutting tests.

This study establishes the utility of torsional vibration, time-frequency analysis, and neural networks for on-line estimation of the extent of flank wear in a milling insert. First, a time-frequency distribution, i.e., a Choi-Williams distribution, is calculated from the torsional vibration of a milling machine spindle. Second, scattering matrices and orthogonalization are employed to identify the time-frequency components that are best correlated to the extent of wear. Third, a neural network is trained to estimate the extent of wear from these critical time-frequency components. To reduce the cost of the system, a nonlinear model-based virtual sensor is developed to eliminate the need for a permanent torsional vibration sensor, which is more expensive and bulky than the translational vibration sensor. Experimental results have shown the usefulness of the proposed solution.