ABSTRACT
The calibration of constitutive models is based on the solution of an optimization problem, where parameter values are sought such that optimal correspondence between measured and simulated response is achieved. The constitutive model studied was especially developed for elastomers with non-linear rate-dependent behaviour. This chapter discusses the calibration of the Bergström-Boyce model for a silicone rubber material and outlines the optimization strategy as well as the sensitivity assessment. The finite element-format of the calibration problem is given using a discontinuous-Galerkin formulation, and the gradient-based solution algorithm is derived. The sensitivities in the parameter values are assessed from assumed noise levels in the data, whereby a method based on the "dual problem" is adopted. In the preliminary study the objective function was found to be "flat" in the parameter space and; hence, the obtained parameter values are not completely reliable. The experimental data used are from uniaxial compression of an ASTM sized specimen of silicone rubber at room temperature.
