skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Modeling and Simulation of High Dimensional Stochastic Multiscale PDE Systems at the Exascale

Abstract

The thrust of the proposal was to exploit modern data-mining tools in a way that will create a systematic, computer-assisted approach to the representation of random media -- and also to the representation of the solutions of an array of important physicochemical processes that take place in/on such media. A parsimonious representation/parametrization of the random media links directly (via uncertainty quantification tools) to good sampling of the distribution of random media realizations. It also links directly to modern multiscale computational algorithms (like the equation-free approach that has been developed in our group) and plays a crucial role in accelerating the scientific computation of solutions of nonlinear PDE models (deterministic or stochastic) in such media – both solutions in particular realizations of the random media, and estimation of the statistics of the solutions over multiple realizations (e.g. expectations).

Authors:
 [1]
  1. Princeton Univ., NJ (United States)
Publication Date:
Research Org.:
Princeton Univ., NJ (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1347551
Report Number(s):
DOE-PRINCETON-2601
DOE Contract Number:
SC0005176
Resource Type:
Technical Report
Resource Relation:
Related Information: ITEM A: Kevrekidis_ER2602_DOE_Final_Report.pdf
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; heterogeneity; uncertainty; complex systems; microscopic dynamics; scientific computation; network dynamics; data mining

Citation Formats

Kevrekidis, Ioannis. Modeling and Simulation of High Dimensional Stochastic Multiscale PDE Systems at the Exascale. United States: N. p., 2017. Web. doi:10.2172/1347551.
Kevrekidis, Ioannis. Modeling and Simulation of High Dimensional Stochastic Multiscale PDE Systems at the Exascale. United States. doi:10.2172/1347551.
Kevrekidis, Ioannis. Wed . "Modeling and Simulation of High Dimensional Stochastic Multiscale PDE Systems at the Exascale". United States. doi:10.2172/1347551. https://www.osti.gov/servlets/purl/1347551.
@article{osti_1347551,
title = {Modeling and Simulation of High Dimensional Stochastic Multiscale PDE Systems at the Exascale},
author = {Kevrekidis, Ioannis},
abstractNote = {The thrust of the proposal was to exploit modern data-mining tools in a way that will create a systematic, computer-assisted approach to the representation of random media -- and also to the representation of the solutions of an array of important physicochemical processes that take place in/on such media. A parsimonious representation/parametrization of the random media links directly (via uncertainty quantification tools) to good sampling of the distribution of random media realizations. It also links directly to modern multiscale computational algorithms (like the equation-free approach that has been developed in our group) and plays a crucial role in accelerating the scientific computation of solutions of nonlinear PDE models (deterministic or stochastic) in such media – both solutions in particular realizations of the random media, and estimation of the statistics of the solutions over multiple realizations (e.g. expectations).},
doi = {10.2172/1347551},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed Mar 22 00:00:00 EDT 2017},
month = {Wed Mar 22 00:00:00 EDT 2017}
}

Technical Report:

Save / Share:
  • Predictive Modeling of multiscale and Multiphysics systems requires accurate data driven characterization of the input uncertainties, and understanding of how they propagate across scales and alter the final solution. This project develops a rigorous mathematical framework and scalable uncertainty quantification algorithms to efficiently construct realistic low dimensional input models, and surrogate low complexity systems for the analysis, design, and control of physical systems represented by multiscale stochastic PDEs. The work can be applied to many areas including physical and biological processes, from climate modeling to systems biology.
  • Acceleration driven instabilities of fluid mixing layers include the classical cases of Rayleigh-Taylor instability, driven by a steady acceleration and Richtmyer-Meshkov instability, driven by an impulsive acceleration. Our program starts with high resolution methods of numerical simulation of two (or more) distinct fluids, continues with analytic analysis of these solutions, and the derivation of averaged equations. A striking achievement has been the systematic agreement we obtained between simulation and experiment by using a high resolution numerical method and improved physical modeling, with surface tension. Our study is accompanies by analysis using stochastic modeling and averaged equations for the multiphase problem.more » We have quantified the error and uncertainty using statistical modeling methods.« less
  • Since the events at the Fukushima-Daiichi nuclear power plant in March 2011 significant research has unfolded at national laboratories, universities and other institutions into alternative materials that have potential enhanced accident tolerance when compared to traditional \uo~fuel zircaloy clad fuel rods. One of the potential replacement fuels is uranium silicide (\usi) for its higher thermal conductivity and uranium density. The lower melting temperature is of potential concern during postulated accident conditions. Another disadvantage for \usi~ is the lack of experimental data under power reactor conditions. Due to the aggressive development schedule for inserting some of the potential materials into leadmore » test assemblies or rods by 2022~\cite{bragg-sitton_2014} multiscale multiphysics modeling approaches have been used to provide insight into these materials. \\ \noindent The purpose of this letter report is to highlight the multiscale modeling effort for \usi~fuel as part of the Nuclear Energy Advanced Modeling and Simulation (NEAMS) Accident Tolerant Fuel (ATF) High Impact Problem (HIP). This fiscal year two new models for \usi~fuel have been incorporated into the BISON fuel performance code~\cite{Williamson2012} based upon lower length scale simulations using the GRASS-SST rate theory code. The models are briefly described in the following sections.« less
  • Since the events at the Fukushima-Daiichi nuclear power plant in March 2011 significant research has unfolded at national laboratories, universities and other institutions into alternative materials that have potential enhanced ac- cident tolerance when compared to traditional UO2 fuel zircaloy clad fuel rods. One of the potential replacement claddings are iron-chromium-alunimum (FeCrAl) alloys due to their increased oxidation resistance [1–4] and higher strength [1, 2]. While the oxidation characteristics of FeCrAl are a benefit for accident tolerance, the thermal neu- tron absorption cross section of FeCrAl is about ten times that of Zircaloy. This neutronic penalty necessitates thinner cladding. Thismore » allows for slightly larger pellets to give the same cold gap width in the rod. However, the slight increase in pellet diameter is not sufficient to compensate for the neutronic penalty and enriching the fuel beyond the current 5% limit appears to be necessary [5]. Current estimates indicate that this neutronic penalty will impose an increase in fuel cost of 15-35% [1, 2]. In addition to the neutronic disadvantage, it is anticipated that tritium release to the coolant will be larger because the permeability of hydrogen in FeCrAl is about 100 times higher than in Zircaloy [6]. Also, radiation-induced hardening and embrittlement of FeCrAl need to be fully characterized experimentally [7]. Due to the aggressive development schedule for inserting some of the potential materials into lead test assemblies or rods by 2022 [8] multiscale multiphysics modeling approaches have been used to provide insight into these the use of FeCrAl as a cladding material. The purpose of this letter report is to highlight the multiscale modeling effort for iron-chromium-alunimum (FeCrAl) cladding alloys as part of the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program through its Accident Tolerant Fuel (ATF) High Impact Problem (HIP). The approach taken throughout the HIP is to utilize lower length scale approaches (e.g., density functional theory, cluster dynamics, rate theory, phase field, and Visco-Plastic- Self-Consistent (VPSC)) to develop more physically informed models at the engineering scale for use in the BISON [9] fuel performance code.« less
  • In this white paper, a road map is presented to establish a multiphysics simulation capability for the design and optimization of sensor systems that incorporate nanomaterials and technologies. The Engineering Directorate's solid/fluid mechanics and electromagnetic computer codes will play an important role in both multiscale modeling and integration of required physics issues to achieve a baseline simulation capability. Molecular dynamic simulations performed primarily in the BBRP, CMS and PAT directorates, will provide information for the construction of multiscale models. All of the theoretical developments will require closely coupled experimental work to develop material models and validate simulations. The plan ismore » synergistic and complimentary with the Laboratory's emerging core competency of multiscale modeling. The first application of the multiphysics computer code is the simulation of a ''simple'' biological system (protein recognition utilizing synthesized ligands) that has a broad range of applications including detection of biological threats, presymptomatic detection of illnesses, and drug therapy. While the overall goal is to establish a simulation capability, the near-term work is mainly focused on (1) multiscale modeling, i.e., the development of ''continuum'' representations of nanostructures based on information from molecular dynamics simulations and (2) experiments for model development and validation. A list of LDRDER proposals and ongoing projects that could be coordinated to achieve these near-term objectives and demonstrate the feasibility and utility of a multiphysics simulation capability is given.« less