Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 28 - Computation, Design, and Quality Assurance of Physical Science and Engineering Applications
Type: Contributed
Date/Time: Sunday, August 8, 2021 : 1:30 PM to 3:20 PM
Sponsor: Section on Physical and Engineering Sciences
Abstract #318933
Title: Latent Function Calibration to Estimate Shock Hugoniots
Author(s): Scott Vander Wiel*
Companies: Los Alamos National Laboratory
Keywords: regularization; physical model; uncertainty quantification
Abstract:

A Hugoniot is the set of thermodynamic states that can be achieved by shocking a material and is an important entity for simulating performance under extreme conditions. I describe fitting a physical shock model to experimental data on porous CeO2. The physical model includes a nonparametric latent function relating the characteristic pressure of the material to its degree of porosity. This function is estimated with a penalized spline technique. Uncertainties are then forward propagated to Hugoniot estimates. The statistical modeling has several interesting features: (i) estimation of systematic errors beyond reported measurement uncertainties, (ii) a local-linear approximation to the nonlinear measurement error likelihood, and (iii) use of an emulator for Bayesian calibration of the physical shock model.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program