Abstract:With the continuous increase of computational capacity, more and more complex nonlinear elastoplastic constitutive models were developed to study the mechanical behavior of elastoplastic materials. These constitutive models generally contain a large amount of physical and phenomenological parameters, which often require a large amount of computational costs to determine. In this paper, an inverse parameter determination method is proposed to identify the constitutive parameters of elastoplastic materials, with the consideration of both strain rate effect and temperature effect. To carry out an efficient design, a hybrid optimization algorithm that combines the genetic algorithm and the Nelder-Mead simplex algorithm is proposed and developed. The proposed inverse method was employed to determine the parameters for an elasto-viscoplastic constitutive model and Johnson-cook model, which demonstrates the capability of this method in considering strain rate and temperature effect, simultaneously. This hybrid optimization algorithm shows a better accuracy and efficiency than using a single algorithm. Finally, the predictability analysis using partial experimental data is completed to further demonstrate the feasibility of the proposed method.

An inverse determination method for strain rate and temperature dependent constitutive model of elastoplastic materials