Uncertainty in Yield Stress Coefficient Identification in Two Steels with Different Machinability. Influence on the Modelling of the Chip Formation Process by Finite Elements

Abstract

Chip Formation process is a complex process, although significant improvements have been reached in the last years, some issues, such as serrated chip formation or residual stresses, remain still obscure. Modelling the chip formation process by finite elements allow us to take into account the non‐linear behaviour of the cutting phenomenon, and hence get information about some variables such as temperature, plastic strain and strain rate that are complicated to be measured by experimental means. There is no doubt that this sort of modelling is making improvements in many fields of machining, e.g., in tools development, cutting conditions selection, but there are some obstacles in its development: (i) the identification of input parameters, (ii) set up of the numerical models and (iii) experimental validation. This paper studies the influence of input parameters in numerical modelling of the chip formation process and particularly the law of workpice behaviour. Two steels with similar composition but different machinability rate have been identified in order to obtain the yield stress coefficient and its uncertainty. Using sensitivity analysis developed in Abaqus/Explicit and proposed in previous works, we get an estimation of the error as a result of the material identification uncertainty. We conclude that it is possible to determine the capability of modelling by finite elements for distinguishing steels with different machinability.