Regression Based Neural Network Modeling for Forecasting of the Metal Volume Removal Rate in Turning Operations

Abstract

The present paper focuses on two techniques, namely regression and neural network, for predicting tool wear. Predicted values of tool wear by both techniques were compared with experimental values. Also, the effects of the main machining variables on tool wear have been determined. The metal volume removed (MVR) was taken as response (output) variable and cutting speed, feed rate, depth of cut and hardness were taken as input parameters, respectively. The relationship between tool wear and machining parameters was found out by direct measurement of the tool wear by MVR. The results showed the ability of regression and neural network models to predict the tool wear, accurately.