Inverse Identification of Johnson-Cook Material Parameters from Machining Simulations
A material model is a prerequisite to the modelling of machining processes. Owing to its versatility, the Johnson-Cook model is commonly used for machining simulations. Determination of the model parameters from experiments is challenging due to the large variations of strains, strain-rates and temperatures which lead to several problems. State-of-the-art experimental methods have to rely on data obtained from much lower strains and strain-rates than those encountered during machining. In this paper, an inverse method of identifying Johnson-Cook parameters from machining experiments is described. A fnite-element model of the machining process was created and a particular Johnson-Cook parameter set was taken from literature for the simulation. The Levenberg-Marquardt Algorithm was used to re-identify the material parameters by looking at the Chip-morphology and the Cutting force evolution. It is shown that the optimisation parameters and error function must be chosen carefully in order to achieve better solutions at lower computational expense.