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

Title: A Supervised Learning Framework for Arbitrary Lagrangian-Eulerian Simulations

; ; ;
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
OSTI Identifier:
Report Number(s):
DOE Contract Number:
Resource Type:
Resource Relation:
Conference: Presented at: International Conference on Machine Learning Applications, Anaheim, CA, United States, Dec 18 - Dec 20, 2016
Country of Publication:
United States

Citation Formats

Jiang, M, Gallagher, B, Kallman, J, and Laney, D. A Supervised Learning Framework for Arbitrary Lagrangian-Eulerian Simulations. United States: N. p., 2016. Web.
Jiang, M, Gallagher, B, Kallman, J, & Laney, D. A Supervised Learning Framework for Arbitrary Lagrangian-Eulerian Simulations. United States.
Jiang, M, Gallagher, B, Kallman, J, and Laney, D. 2016. "A Supervised Learning Framework for Arbitrary Lagrangian-Eulerian Simulations". United States. doi:.
title = {A Supervised Learning Framework for Arbitrary Lagrangian-Eulerian Simulations},
author = {Jiang, M and Gallagher, B and Kallman, J and Laney, D},
abstractNote = {},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 2016,
month = 7

Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • The analysis of fluid-structure interaction involves the calculation of both fluid transient and structure dynamics. In the structural analysis, Lagrangian meshes have been used exclusively, whereas for the fluid transient, Lagrangian, Eulerian, and arbitrary Lagrangian-Eulerian (quasi-Eulerian) meshes have been used. This paper performs an evaluation on these three types of meshes. The emphasis is placed on the applicability of the method in analyzing fluid-structure interaction problems in HCDA analysis.
  • Machine learning is used in many applications, from machine vision to speech recognition to decision support systems, and is used to test applications. However, though much has been done to evaluate the performance of machine learning algorithms, little has been done to verify the algorithms or examine their failure modes. Moreover, complex learning frameworks often require stepping beyond black box evaluation to distinguish between errors based on natural limits on learning and errors that arise from mistakes in implementation. We present a conceptual architecture, failure model and taxonomy, and failure modes and effects analysis (FMEA) of a semi-supervised, multi-modal learningmore » system, and provide specific examples from its use in a radiological analysis assistant system. The goal of the research described in this paper is to provide a foundation from which dependability analysis of systems using semi-supervised, multi-modal learning can be conducted. The methods presented provide a first step towards that overall goal.« less
  • A Three-Dimensional Finite Volume Arbitrary Lagrangian-Eulerian simulation code was developed to study different plasma physics problems in 3D+t. The code is based on a complex multi-component species program with transport and radiation terms written and applied to plasma and fusion physics problems. Three different examples are shown: double-base chemical propellant combustion, ignition and propagation of a thermonuclear detonation wave, and, the development of the Kelvin-Helmholtz (KH) instability in local plane slab models of the magnetopause, showing the response of a background equilibrium to the excitation by finite amplitude perturbations generated upstream.
  • A description is given of an arbitrary Lagrangian-Eulerian method for analyzing fluid-structure interactions in fast reactor containment with complex internal structures. In the analysis, the fluid transient can be calculated either implicitly or explicitly, using a finite-difference mesh with vertices that may be moved with the fluid (Lagrangian), held fixed (Eulerian), or moved in any other prescribed manner (hybrid Lagrangian-Eulerian). The structural response is computed explicitly by two nonlinear, elastic-plastic finite element modules formulated in corotational coordinates. Interaction between fluid and structure is accounted for by enforcing the interface boundary conditions. The method has convincing advantages in treating complicated phenomenamore » such as flow through perforated structures, large material distortions, flow around corners and irregularities, and highly contorted fluid boundaries. Several sample problems are given to illustrate the effectiveness of this arbitrary Lagrangian-Eulerian method. 12 refs.« less
  • In this study we provided an experimental test bed for validating features of the Arbitrary Lagrangian Eulerian Grid for Research Applications (ALEGRA) code over a broad range of strain rates with overlapping diagnostics that encompass the multiple responses. A unique feature of the ALEGRA code is that it allows simultaneous computational treatment, within one code, of a wide range of strain-rates varying from hydrodynamic to structural conditions. This range encompasses strain rates characteristic of shock-wave propagation (107/s) and those characteristics of structural response (102/s). Most previous code validation experimental &udies, however, have been restricted to simulating or investigating a singlemore » strain-rate regime. What is new and different in this investigation is that we have performed well-controlled and well-instrumented experiments, which capture features relevant to both hydrodynamic and structural response in a single experiment. Aluminum was chosen for use in this study because it is a well-characterized material. The current experiments span strain rate regimes of over 107/s to less than 102/s in a single experiment. The input conditions were extremely well defined. Velocity interferometers were used to record the high' strain-rate response, while low strain rate data were collected using strain gauges. Although the current tests were conducted at a nominal velocity of - 1.5 km/s, it is the test methodology that is being emphasized herein. Results of a three-dimensional experiment are also presented.« less