Bayesian calibration of strength parameters using hydrocode simulations of symmetric impact shock experiments of Al-5083

Walters, Donald L.; Biswas, Ayan; Lawrence, Earl; Francom, Devin; Luscher, Darby J.; Fredenburg, David A.; Moran, Kelly R.; Sweeney, Christine; Sandberg, Richard L.; Ahrens, James; Bolme, Cynthia

doi:10.1063/1.5051442

Cited by 44 publications

(15 citation statements)

References 35 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. Bayesian calibration frameworks are being actively developed specifically for these types of dynamic experiments in which velocimetry is the primary diagnostic and a predictive simulation capability exists [29,30]. There are two features of Bayesian approaches which are appealing for this application.…”

Section: Bayesian Calibrationmentioning

confidence: 99%

Multi-megabar Dynamic Strength Measurements of Ta, Au, Pt, and Ir

Brown

Davis

Seagle

2020

J. dynamic behavior mater.

View full text Add to dashboard Cite

Magnetic loading was used to shocklessly compress four different metals to extreme pressures. Velocimetry monitored the behavior of the material as it was loaded to a desired peak state and then decompressed back down to lower pressures. Two distinct analysis methods, including a wave profile analysis and a novel Bayesian calibration approach, were employed to estimate quantitative strength metrics associated with the loading reversal. Specifically, we report for the first time on strength estimates for tantalum, gold, platinum, and iridium under shockless compression at strain rates of ∼ 5 × 10 5 /s in the pressure range of ∼ 100-400 GPa. The magnitude of the shear stresses supported by the different metals under these extreme conditions are surprisingly similar, representing a dramatic departure from ambient conditions.

show abstract

Section: Bayesian Calibrationmentioning

confidence: 99%

Multi-megabar Dynamic Strength Measurements of Ta, Au, Pt, and Ir

Brown

Davis

Seagle

2020

J. dynamic behavior mater.

View full text Add to dashboard Cite

show abstract

“…The purple lines indicate the results of the associated physical experiments that will be used as part of calibrating the simulation model. For more details of this example, see Walters et al [28].…”

Section: Flyer Platementioning

confidence: 99%

A comparison of Gaussian processes and neural networks for computer model emulation and calibration

Myren

Lawrence

2021

Statistical Analysis

View full text Add to dashboard Cite

The Department of Energy relies on complex physics simulations for prediction in domains like cosmology, nuclear theory, and materials science. These simulations are often extremely computationally intensive, with some requiring days or weeks for a single simulation. In order to assure their accuracy, these models are calibrated against observational data in order to estimate inputs and systematic biases. Because of their great computational complexity, this process typically requires the construction of an emulator, a fast approximation to the simulation. In this paper, two emulator approaches are compared: Gaussian process regression and neural networks. Their emulation accuracy and calibration performance on three real problems of Department of Energy interest is considered. On these problems, the Gaussian process emulator tends to be more accurate with narrower, but still well-calibrated uncertainty estimates.The neural network emulator is accurate, but tends to have large uncertainty on its predictions. As a result, calibration with the Gaussian process emulator produces more constrained posteriors that still perform well in prediction.

show abstract

“…The RMI experiments were modeled using FLAG, an arbitrary Lagrangian-Eulerian, explicit, finite-volume hydrodynamics code using arbitrary polyhedral computational zones (elements) [40][41][42][43][44], which has been used for modeling many physical phenomena [45][46][47][48][49][50][51]. No mesh relaxation or Eulerian remapping was used in these calculations because all the spikes arrested in these RMI experiments leading to only modest deformations.…”

Section: Analysis Approach and Modelingmentioning

confidence: 99%

Tantalum strength at extreme strain rates from impact-driven Richtmyer-Meshkov instabilities

et al. 2019

View full text Add to dashboard Cite

Recently, Richtmyer-Meshkov instability (RMI) experiments driven by high explosives and fielded with perturbations on a free surface have been used to study strength at extreme strain rates and near zero pressure. The RMI experiments reported here used impact loading, which is experimentally simpler, more accurate to analyze, and which also allows the exploration of a wider range of conditions. Three experiments were performed on tantalum at shock stresses from 20 to 34 GPa, with six different perturbation sizes at each shock level, making this the most comprehensive set of strength-focused RMI experiments reported to date on any material. The resulting estimated average strengths of 1200-1400 MPa at strain rates of 10 7 /s exceeded, by 40% or more, a common power law extrapolation from data at strain rates below 10 4 /s. Taken together with other data in the literature that show much higher strength at simultaneous high rates and high pressure, these RMI data isolated effects and indicated that, in the range of conditions examined, the pressure effects are more significant than rate effects.

show abstract

Bayesian calibration of strength parameters using hydrocode simulations of symmetric impact shock experiments of Al-5083

Abstract: Dynamic evolution of microstructure during laser shock loading and spall failure of single crystal Al at the atomic scales

Cited by 44 publications

References 35 publications

Multi-megabar Dynamic Strength Measurements of Ta, Au, Pt, and Ir

Multi-megabar Dynamic Strength Measurements of Ta, Au, Pt, and Ir

A comparison of Gaussian processes and neural networks for computer model emulation and calibration

Tantalum strength at extreme strain rates from impact-driven Richtmyer-Meshkov instabilities

Contact Info

Product

Resources

About