Parameter identification for viscoplastic models based on analytical derivatives of a least-squares functional and stability investigations (Q677043)

The paper deals with the identification of material parameters of viscoplastic constitutive equations from uniaxial test data, when the deformations are small. In mathematical modelling of uniaxial viscoplastic problems, the constitutive equations for inelastic part of deformation and for a set of internal-nonmeasurable variables are given in a rate form dependent on uniaxial stress, on temperature, and on a set of material parameters. The direct and inverse problems are formulated in stress controlled one-dimensional viscoplasticity. In order to have a well-posed problem, the backward calculation is replaced by an optimal approach strategy. In many situations this problem may lead to numerically instable solutions. The discretized formulations of the direct and inverse problems are presented. The authors discuss solution strategies based on linearization procedures where a recursion formula is obtained for the inverse problem, and outline the optimization strategies based on gradient evaluations. Some specific issues are put into evidence: indefinite Hessians, line-search strategies, scaling and use of different iteration matrices. Numerical results are performed for monotonic and cyclic uniaxial tension and compression tests. Comparisons with a generic algorithm reflect the efficiency of the proposed strategy with respect to execution time. The effect of perturbations of the experimental data on the stability of parameters is also analysed.