Browse Prior Art Database

Integrated Solution Visualization, Validation, and Steering for Retinable Finite Element Modeling Results

IP.com Disclosure Number: IPCOM000117977D
Original Publication Date: 1996-Aug-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 2 page(s) / 98K

Publishing Venue

IBM

Related People

Bala, GP: AUTHOR [+3]

Abstract

Disclosed is a method that enables the visualization, validation, and steering of intermediate results produced by Finite Element Modeling (FEM) solvers possessing automatic solution refinement capabilities. The novelty of the approach is the sharing of the computational data and functional resources within the core of the solver by the visualization kernel. In addition, integration is further tightened by allowing the visualization procedure the ability to yield control back to the processor when a need arises to update the model and simulation parameters (e.g., boundary conditions) during an analysis.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Integrated Solution Visualization, Validation, and Steering for Retinable
Finite Element Modeling Results

      Disclosed is a method that enables the visualization,
validation, and steering of intermediate results produced by Finite
Element Modeling (FEM) solvers possessing automatic solution
refinement capabilities.  The novelty of the approach is the sharing
of the computational data and functional resources within the core of
the solver  by the visualization kernel.  In addition, integration is
further tightened by allowing the visualization procedure the ability
to yield  control back to the processor when a need arises to update
the model and  simulation parameters (e.g., boundary conditions)
during an analysis.

The steps in the procedure are shown in the Figure and detailed in
the following:
  1.  In this first step, the model geometry and the simulation
       parameters (e.g., material properties, boundary conditions,
       etc.) are collected from the preprocessor.
  2.  Next, a solution is computed using the finite element solver.
  3.  The solution is then automatically examined, via the data
       structures and computational function shared with the solver,
       to determine the way in which it will be sampled.  This
       determination is based upon the numerical behavior of the
       solution and upon the current viewing characteristics
       (e.g., sizes of features relative to the viewing scope,
       screen resolution, interactively-specified sampling
       parameters, etc.).
  4.  Given the resolution determined above, the solution data are
       queried at the resulting sample points.  Exact numerical data
       can also be displayed, on demand, at any location on the
       model geometry where the solution has been computed.
  5.  The sample data are then displayed using the display feature
       available on the target graphics platform (e.g., continuous
       tone shaded surface plots, contour lines, deformations,
       vector plots, etc.).
  6.  At this point, it is determined if refinement of the
       solution will be needed.
  7.  If such refinement is required, the refinement parameters
       (e.g., tolerances, mesh densities, etc.) may be updated.
  8.  The next refinement is then computed.
  9.  If, instead of updating any refinement parameters,...