Parameter identification for a Cowper-Symonds material model using a genetic algorithm combined with a response surface

Abstract

Purpose In conditions where a product is subjected to extreme mechanical loading in a very short time, a strain rate has a significant influence on the behaviour of the product’s material. To accurately simulate the behaviour of the material during these loading conditions, the strain rate parameters of the selected material model should be appropriately used. This paper aims to present a fast method with which the proper strain-rate-dependent parameter values of the selected material model can be easily determined. Design/methodology/approach In the paper, an experiment was designed to study the behaviour of thin, flat, metal sheets during an impact. The results from this experiment were the basis for the determination of the strain-rate-dependent parameter values of the Cowper–Symonds material model. Optimisation processes with different numbers of required parameters of the selected material model were performed. The optimisation process consists of the method for design of experiment, modelling a response surface and a genetic algorithm. Findings The paper provides comparison of two optimisation processes with different methods for design of experiment. The performances of the presented method are compared and the engineering applicability of the results is discussed. Originality/value This paper presents a new fast approach for the identification of the parameter values of the Cowper–Symonds material model, if these cannot be easily determined directly from experimental data.