Neural Network Material Modelling for Machining FEM Simulation

Although the correct determination of the material constitutive properties is the key to the successful FEM simulation of machining, the material rheological behaviour at the high temperature, strain rate, and strain conditions encountered in chip formation cannot be provided by normal stress-strain curves or impact tests. In this paper, a neural network approach is used to model the flow dynamics work material behaviour by reconstructing the stress-strain curves of carbon steel work material from tensile test experimental data.