Inverse analysis procedure for identifying material parameters of hypoplastic constitutive model for soils

In this paper we present an inverse analysis method for material parameter identification applied to the hypoplastic constitutive model for soils. The proposed method is classified as a direct-approach one, since the discrepancy between the material responses, measured and calculated by means of the finite element method is directly minimized. In the presented work this minimization is made up of the modified Levenberg-Marquardt optimization method based on the so-called trust region approach. We provide a numerical example of parameter identification for a triaxial test performed on a sand material, supplemented by the analysis of the influence of the initial guesses of the parameters to be identified, as well as the errors in the measured control data on the identification process.