uncertainty quantification

Dear Members of the Mechanics and Materials Communities,

You may have known about the efforts of the Mechanics of Materials and Structures (MoMS) program of the National Science Foundation on revitalizing the practice of verification and validation (V&V) of computational models as a rigorous and indispensable step in the scientific investigative process.

Esteemed Colleagues,

Dear Colleagues,

As part of the IMECE 2020 (November 13-19, 2020, Portland, Oregon), we are organizing a topic on “Data-Enabled Predictive Modeling, Machine Learning, and Uncertainty Quantification in Computational Mechanics.” It is listed in Track 12: Mechanics of Solids, Structures, and Fluids: https://event.asme.org/IMECE/Program/Tracks-Topics.

We've recently published our new study about Uncertainty Quantification in Molecular Dynamics (MD) Simulations. Due to the selection of functional forms of interatomic potentials or the numerical approximation, MD simulations may predict different material behavior from experiments or other high-fidelity results. In this study, we used Stochastic Reduced Order Modeling (SROM) to achieve

(1) mechanical behavior of graphene predicted by MD simulations in good agreement with the continuum model which has been calibrated by experiments;

Dear colleagues,

We encourage you to submit your abstracts to minisymposium 1801 of the 13th World Congress on Computational Mechanics (New York City, from July 22 to 27 of 2018). This minisymposium focuses on:

1. recently developed methods for data-driven approaches;

2. data-driven applications to fluids, structures and materials involving (but not limited to) machine learning, uncertainty quantification and/or optimization.

Several positions at the PhD and postdoctoral levels are available in the group of Prof. Guilleminot at Duke University. The ideal candidates should have a particular interest in conducting interdisciplinary research at the interface of computational mechanics, materials science and uncertainty quantification. Topics of interest include (but are not limited to) stochastic modeling and computational frameworks for large-scale nonlinear materials and systems, statistical inverse problems and multiscale approaches.

Reference : Accelerating Monte Carlo estimation with derivatives of high-level finite element models
http://orbilu.uni.lu/handle/10993/28618

Bayesian inference for the stochastic identification of elastoplastic material parameters: Introduction, misconceptions and insights http://orbilu.uni.lu/handle/10993/28631