Abstract
The paper presents the application of plane strain compression (PSC) test to the identification of material model parameters. Moreover, an attempt to substitute the finite element (FE) model with the metamodel in the inverse analysis was made. There are several plastometric tests carried out in order to obtain material properties. Despite the fact that PCS test is very accurate, the most commonly used is uniaxial compression (UC) test. This is due to the difficulties in interpretation of results of the PSC test. Inverse analysis is used to overcome this problem. This analysis requires simulation results of adopted test. FE model is most commonly used in these simulations. However, due to mesh density in the area of greatest strain gradients, FE simulations can take unacceptably long time. Therefore, artificial neural network (ANN) was proposed in the present work as metamodel of the PSC test. This metamodel was used in the inverse analysis performed to obtain material properties for copper alloys. The results were compared with uniaxial compression test performed for the same material and good agreement was observed.