Constitutive Model Calibration via Autonomous Multiaxial Experimentation

Abstract

Modern plasticity models contain numerous parameters that can be difficult and time consuming to fit using current methods. Additional experiments are seldom conducted to validate the model for experimental conditions outside those used in the fitting procedure. To increase the accuracy and validity of these advanced constitutive models, software and testing methodology have been developed to seamlessly integrate experimentation, parameter identification, and model validation in real-time over a range of multiaxial stress conditions, using an axial/torsional test machine. Experimental data is reduced and finite element simulations are conducted in parallel with the test based on experimental strain conditions. Optimization methods reconcile the experiment and simulation through changes to the plasticity model parameters. Excursions into less-traveled portions of the multiaxial stress space can be predicted, and then executed experimentally, to identify deficiencies in the model. Most notably, the software can autonomously redirect the experiment to increase the robustness of the plasticity model where further deficiencies are identified, thus providing closed loop control of the experiment. This novel process yields a calibrated plasticity model upon test completion that has been fit and more importantly validated, and can be used directly in finite element simulations of more complex geometries.

This paper is declared a work of the U.S. Government and is not subject to copyright protection in the United States.