On the interest of using full field measurements in ductile damage model calibration

This paper is focused on the identification method by inverse analysis of ductile damage model parameters. The Lemaitre damage model is considered. The aim is to assess the interest of using local measurements, such as full field measurements, in the calibration process. Basic approach using only load-displacement measurements from a tensile test is first described and discussed. Secondly necking measurements are added in the calibration process. And finally an inverse analysis approach including full field measurements in a large deformation framework is presented. For each step particular attention is paid to the definition of the objective function to catch efficiently the softening behavior and the fracture of the sample. This step-by-step approach allows understanding the lack of information embedded in the load-displacement curve to calibrate such a model. It also highlights the multi-extrema aspect of the optimization problem. The introduction of local measurements, as necking and even more as displacement fields, is an efficient way of overcoming this lack of information. The analysis is based on kriging response-surfaces built during the minimization of the objective function. These surfaces highlight the weak sensitivity of one of the Lemaitre model parameters (the b parameter) when only the load-displacement curve is taken into account. The sensitivity is then improved when local measurements are considered. The use of local measurement also leads to the removal of correlation between model parameters.