Prediction of cutting force during hard turning of 105WCr6 steel using artificial neural network and neuro-fuzzy modeling

Abstract

In this work research of correlation between mean value and range of cutting force and processing modes during hard turning of 105WCr6 steel is presented. The results of three-factor experiment on end face cutting of ring workpieces hardened to 55 HRC are presented. During experiment cutting speed, feed and cutting depth are varied. The value of the cutting force is estimated indirectly by the value of current load of the main drive motor. For the development of the model which can predict the value of cutting force at given cutting modes feed-forward neural network trained using Bayesian regularization algorithm and adaptive neuro-fuzzy inference system are used. Developed mathematical models can predict cutting force parameters with high accuracy.