Online Tool Wear Prediction in Milling Operation Using Feed Motor Current Signal

Abstract

Abstract-In the present work, artificial neural network (ANN) and regression analysis are used to predict the real time flank wear of a cutting tool (KC710) in milling operation. The various parameters such as root mean square (RMS) value of the spindle motor current, spindle speed, depth of cut and feed-rate are the inputs to the network, and flank wear is the output. Online data of milling operation is taken for training and testing of the neural networks. Effect of various cutting conditions (speed, depth of cut, feed-rate) on the flank wear has been investigated. The flank wear prediction using the trained network is found to be in better agreement in comparison with the experimentally measured flank wear of the milling tool. The value of co-efficient of determination (R 2 ), between the model predictions and the experimentally obtained value are compared. Accuracy of the predicted flank wear using neural network is found to be better than that using regression model.