Real-Time Prediction of Workpiece Errors for a CNC Turning Centre, Part 4. Cutting-Force-Induced Errors

A neural network method is presented for predicting cutting-force-induced errors in real-time during turning operations based on the estimated cutting forces. Workpiece errors can be considerably affected by the deflections of the machine–workpiece–tool system. A model of the elastic deflections of the machine–workpiece–tool system due to the cutting force in turning developed. A novel radial basis function (RBF) neural network is used to map the relationship between the cutting-force components (radial, axial and tangential) and the consequent dimensional deviation of the finished parts caused by the combined deflections of the machine–workpiece–tool system. Cutting tests were performed on a two-axis CNC turning centre and the experimental results showed that the prediction accuracy of the maximum diameter error of the workpiece was within 15%. The trained RBF neural network was able to predict the cutting force induced error in real-time during turning.