COMPARISON OF THREE METHODS FOR MATERIAL HARDENING PARAMETER IDENTIFICATION UNDER CYCLIC TENSION-COMPRESSION LOADINGS: ROLL LEVELLING CASE STUDY

摘要

The roll levelling is a forming process used to remove the residual stresses and imperfections of metal strips by means of plastic deformations. The process is especially important to avoid final geometrical errors when coils are cold formed or when thick plates are cut by laser. In the last years, and due to the appearance of high strength materials such as Ultra High Strength Steels, machine design engineers are demanding a reliable tool for the dimensioning of the levelling facilities. In response to this demand, Finite Element Analysis is becoming an important technique able to lead engineers towards facilities optimization through a deeper understanding of the process. In this scenario, the accuracy and quality of the simulation results are highly dependent on the accuracy of the implemented material model. During roll levelling process, the sheet metal is subjected to cyclic tensile-compressive deformations, therefore a proper constitutive model which considers the phenomena that occur during cyclic loadings, such as the Bauschinger effect, work hardening and the transient behaviour, is needed. The prediction of all these phenomena which affect the final shape of the product are linked to the hardening rule. In the present paper, the roll levelling simulation of a DP 1000 steel is performed using a combined isotropic-kinematic hardening formulation introduced by Chaboche and Lemaitre due to its simplicity and its ability to predict the Bauschinger effect. The model has been fitted to the experimental curves obtained from a cyclic tension-compression test, which has been performed by means of a special tool developed to avoid the buckling of the specimen during compressive loadings. The model has been fitted using three different material hardening parameter identification methodologies which have been compared.