On the Prediction of Macroscopic Yield Surfaces of a Pearlitic Steel using Multiscale Modeling

Övrigt konferensbidrag, 2013

On the microscale, pearlite consists of hard and brittle cementite lamellae embedded in a ductile ferrite matrix. The cementite lamellae are arranged in colonies within which the lamella orientation is ideally constant. This composite-like constitution, on the microscale, makes pearlitic steels ideally suited for multiscale modeling. In this contribution a three-scale multiscale modeling setup is used to describe the mechanical behav- ior of a pearlitic steel. The macroscale represents the engineering scale on which a typical structural component would be analyzed. The mesoscale comprises colonies, with varying orientations (both mor- phological and crystallographic), thereby enabling the interactions between colonies to be taken into account. On the microscale a model representing the lamellar structure of pearlite is used. This model accounts for the behavior of the constituents but also the interactions between them. A cornerstone in this contribution is the formulation of a macroscopic, energy based, yield criterion based on homogenized quantities (cf. e.g. [1, 2, 3]). With such a criterion macroscopic yield surfaces can be predicted. The impact of altering the prolongation condition on the resulting yield surface is studied. Furthermore, the effect of adding a pre-loading before carrying out the yield surface prediction is investigated. Regarding the topic of how to identify the correct values of the parameters in a multiscale model several possibilities exists. This topic will be discussed briefly.