ROBUST TOOL WEAR ESTIMATION WITH RADIAL BASIS FUNCTION NEURAL NETWORKS

In this paper, a unified method for constructing dynamic models for to ol wear from prior experiments is proposed. The model approximates fla nk and crater wear propagation and their effects on cutting force usin g radial basis function neural networks. Instead of assuming a structu re for the wear model and identifying its parameters, only an approxim ate model is obtained in terms of radial basis functions. The appearan ce of parameters in a linear fashion motivates a recursive least squar es training algorithm. This results in a model which is available as a monitoring tool for online application. Using the identified model, a state estimator is designed based on the upperbound covariance matrix . This filter includes the errors in modeling the wear process, and he nce reduces filter divergence. Simulations using the neural network fo r different cutting conditions show good results. Addition of pseudo n oise during state estimation is used to reflect inherent process varia bilities. Estimation of wear under these conditions is also shown to b e accurate. Simulations performed using experimental data similarly sh ow good results. Finally, experimental implementation of the wear moni toring system reveals a reasonable ability of the proposed monitoring scheme to track flank wear.