Prediction of Torque in Milling By Response Surface Method and Neural Network

Abstract

The present paper discusses the development of the first-order model for predicting the cutting torque in the milling operation of ASSAB 618 stainless steel using coated carbide cutting tools. The first-order equation was developed using response surface method (RSM). The input cutting parameters were the cutting speed, feed rate, radial depth and axial depth of cut. The study found that the predictive model was able to predict torque values close to those readings recorded experimentally with a 95% confident interval. The results obtained from the predictive model were also compared by using multilayer perceptron with back-propagation learning rule artificial neural network. The first-order equation revealed that the feed rate was the most dominant factor which was followed by axial depth, radial depth and cutting speed. The cutting torque value predicted by using Neural Network was in good agreement with that obtained by RSM. This observation indicates the potential use of RSM in predicting cutting parameters thus eliminating the need for exhaustive cutting experiments to obtain the optimum cutting conditions in terms of torque.