engleski

Identifiability of Ludwik’s law parameters depending on the sample geometry via inverse identification procedure

Understanding the engineering component response under prescribed workloads is essential to improve its operating lifetime. Moreover, insight into the material elastoplastic behavior is crucial since it can be described by various material models. Therefore, to determine the elastoplastic material response, mechanical tests with different loading regimes are carried out. The material response can be monitored by advanced measurement methods such as Digital Image Correlation (DIC), which stands out as one of the most widely used optical method. In this study, an FE-based approach was employed. The resulting measurements (i.e. nodal displacements of the FE mesh) form a straightforward connection with numerical simulations. An inverse problem for determining elastoplastic material parameters can be set by coupling FE-based DIC and numerical simulations in a single advanced identification routine. For this study, Finite Element Model Updating (FEMU) was employed. FEMU has proven to be a valuable identification tool used for both uniaxial and for biaxial loading histories. In this work, a study was performed to evaluate the influence of sample geometry on the identified material parameters. Two distinct sample geometries were considered, namely, the central part of the ligament of the first sample was thinned with a radius of 75 mm (i.e. Dogbone sample), whereas the latter is designed with two parallel edges (DIN sample), to evaluate the material parameters sensitivity to sample geometry.

sensitivity analysis ; inverse identification ; full-field measurements ; HARDOX 450 ; sample geometry

nije evidentirano