|About this Abstract
||2019 TMS Annual Meeting & Exhibition
||Computational Approaches for Big Data, Artificial Intelligence and Uncertainty Quantification in Computational Materials Science
||Quantifying Uncertainty in High Strain Rate Materials Strength with Bayesian Inference
||David Rivera, Jason Bernstein, Katie Schmidt, Nathan Barton, Ana Kupresanin, Jeff Florando
|On-Site Speaker (Planned)
The quantification of uncertainty in materials models is playing an increasing role in engineering design. Of particular importance are materials strength models as they form an integral part of the methodology used to predict plastic deformation in structures subject to load. The large number of parameters in such models however, makes calibration and determination of uncertainty challenging. In this work, an approach to uncertainty quantification founded on Bayesian statistics is applied to estimate the uncertainty in calibrating a continuum materials strength model to high strain rate experimental data consisting of Hopkinson bar and Taylor impact tests. The optimization routine is made tractable through the use of a Gaussian process based surrogate model and the underlying uncertainty in materials strength determined from the posterior distribution of the calibration parameters. The model is then applied to predict the deformation of Taylor impact experiments and the propagation of uncertainty demonstrated.
||Planned: Supplemental Proceedings volume