An objective meta-modeling approach for indentation-based material characterization

摘要

Over the last decade, indentation-based material characterization has shifted away from contact zone measurements for identifying the work hardening properties, and has begun to focus on inverse analysis using an experimental and simulated loading-unloading curve instead. Finite element modeling of the indenter-specimen system is a complex task and the solution to the inverse identification problem using the force-displacement curve is not necessarily unique. Also, the precise measurement of the displacement of the indenter tip requires the determination of the indenter frame compliance and indenter tip deformation. In this paper, we revisit the indentation imprint shape, using a novel approach involving image comparison by manifold learning, and show that it is able to outperform the conventional approach for a variety of cases. We illustrate our protocol using a spherical indenter on specimens of three different materials: AISI1095 steel and two aluminum alloys EN AW-2017F and EN AW-5754F. (C) 2017 Elsevier Ltd. All rights reserved.