|About this Abstract
||2016 TMS Annual Meeting & Exhibition
||Computational Methods for Uncertainty Quantification, Model Validation, and Stochastic Predictions
||Data Analysis in Mesoscale Model of Ductile Damage
||Cristina Garcia-Cardona, Marian Anghel, Ricardo Lebensohn
|On-Site Speaker (Planned)
In order to validate the predictions of mesoscale models of deformation and discover relationships between microstructure and ductile damage in polycrystalline aggregates it is necessary to guarantee that the parameters of the model and the initial conditions are consistent with the conditions of the real experiment. This requires the development of tools that can estimate model parameters from measurements as well as reduce the uncertainty in state estimation. We are implementing such a computational framework based on Fast Fourier Transform, direct and adjoint sensitivity analysis formulations and model reduction techniques. We demonstrate the utility of the implemented framework using data from simulated and real experiments, and show how this contributes to improve the performance in model predictions.
||Planned: A print-only volume