Simplified Calibration of an Academic Refined Model for Industrial Use

Material heterogeneity has a considerable effect on spring back. Integrating heterogeneity in a simulation to obtain a more accurate spring back prediction is complex and often requires a significant number of tests and parameters. Whilst in the industry, the necessity of predicting forming process outcomes, e.g. the spring back of the material, is becoming essential, it does inevitably require a performant and easy to calibrate model. At academic level, a compartmentalized model [1] considers the heterogeneity of the material as one of the parameters driving its plastic behaviour. Its ability to describe load-unload cycles with a small number of adjustable parameters has demonstrated [2]. In this paper, we propose to optimize the identification process on an industrial material. The chosen material, a DP 600 steel, already presents high heterogeneity level due to its initial structure. The main objective is to calibrate the compartmentalized model by using only a monotonous tensile test. The identified model is integrated in a finite element code and simulations of a loading/unloading path are then carried out to evaluate the prediction accuracy of the material behaviour at small strain.. This possible new method could help reduce the number of tests required to obtain the required prediction accuracy, especially when a large spring back effect is present. This would allow to achieve more accurate optimization of the process parameters in the bending operation.