Identification of Material Parameters for Inelastic Constitutive Models Using Stochastic Methods

The parameters of a constitutive model are usually identified by minimization of the distance between model response and experimental data. However, measurement failures and differences in the specimens lead to deviations in the determined parameters. In this article we present our results of a study of these uncertainties for two constitutive models of Chaboche‐type. The models differ only by a kinematic hardening variable. It turns out, that the kinematic hardening variable proposed by Haupt, Kamlah, and Tsakmakis yields a better description quality than the one of Armstrong and Frederick. For the parameter optimization as well as for the study of the deviations of the fitted parameters we apply stochastic methods. The available test data result from creep tests, tension‐relaxation tests and cyclic tests performed on AINSI SS316 stainless steel at 600OC. Since the amount of test data is too small for a proper statistical analysis we apply a stochastic simulation technique to generate artificial data which exhibit the same stochastic behaviour as the experimental data. (© 2007 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)