NON-LINEAR REGRESSION APPLICATION IN IDENTIFICATION OF HYPERELASTIC CONSTITUTIVE MODELS OF SOFT TISSUES

conference paper

English

Choice of a convenient constitutive model for computational modelling of soft tissues is an important assumption of model reliability. The optimal shape of constitutive equations depends not only on the type of material behaviour (isotropic or orthotropic, elastic or viscoelatic, etc.) but on the deformation range to be described as well. For example in blood vessel walls, there are mostly 2D or 3D stress states. If a hyperelastic constitutive model is based (because of a lack of experimental data) on uniaxial tension tests only, the results for biaxial stress states can be quite false, because of transversal strains differing from the test. Methods of regression, which are able to improve the model behaviour, and comparison of predictive capability of models are presented in the paper.

non-linear regression, constitutive relations, soft tissues, identification of properties

VAJDÁK, M.

2003

12. 5. 2003

Association for Engineering Mechanics

Svratka, CZ

80-86246-18-3

Engineering Mechanics 2003

¨360

2