The discrete ordinate method development to the transport equation solving' The 3D code RADUGA5'1P a - PowerPoint PPT Presentation

1 / 16
About This Presentation
Title:

The discrete ordinate method development to the transport equation solving' The 3D code RADUGA5'1P a

Description:

The discrete ordinate method development. to the transport equation solving. ... Isotropic conic source. Problem of radiation safety of astronauts and equipment ... – PowerPoint PPT presentation

Number of Views:153
Avg rating:3.0/5.0
Slides: 17
Provided by: pc75164
Category:

less

Transcript and Presenter's Notes

Title: The discrete ordinate method development to the transport equation solving' The 3D code RADUGA5'1P a


1
The discrete ordinate method developmentto the
transport equation solving.The 3D code
RADUGA-5.1(P)and multiprocessors computers
  • O.V.Nikolaeva,
    L.P.Bass
  • Keldysh Institute of Applied Mathematic, Moscow,
    Russia
  • V.S.Kuznetsov
  • Research Scientific Center "Kurchatov Institute",
    Moscow, Russia

T.A.Germogenova,
2
Methods developmentto solve the transport
equationin different applications
  • radiation shielding problems,
  • atmosphere optic problems,
  • biomedicine problems.
  • All methods are included into
  • the code RADUGA-5.1(P)

3
Flexible algorithm
  • In each subregion grids and scheme are chosen in
    dependence of solution properties in this
    subregion
  • smooth solutions,
  • non-smooth solutions,
  • discontinuous solutions.

4
Calculation of discontinuous solutionby DD and
SWDD schemes
  • Step of spatial mesh 1/27 mfp.

5
Phase functions representation by Legenders
polynomials
PN expansion for a regular component is
used Here is Legenders polynomial.
  • Peak-forward phase function is presented by sum
    regular component and singular one

6
Comparison of numerical resultsfor two methods
of phase function presentation
  • Method 1 by its values at nodes of some grid,
  • Method 2 by Legenders polynomials

Spatial mesh 15 ? 30 meshes Mesh of spatial
grid 0.041 mfp. Angular quadrature S50
7
Solution at axis z
8
Definition of geometry region
  • Boundaries of materials are determined by
    surfaces of simple 2D-3D bodies
  • parallelepiped
  • truncated cone
  • prism
  • cylinder
  • regular hexahenron
  • sphere
  • rectangle
  • triangle
  • circle or sector
  • regular hexagon
  • Each body is defined by its geometrical parameters

Circle approximation by edges of regular spatial
grid meshes
9
Point source of great apertureSkyshine problem
Analytical calculation of un-scattered radiation
intensity
  • Energy interval (15Mev, 0 Mev)
  • 22 energetic groups
  • P3 expansion for phase functions
  • Spatial grid 480 ? 132 meshes
  • Angular quadrature S16

10
Parallel beamSolar radiation reflection by a
cloudof irregular structure.The international
Intercomparison of 3D Radiation Codes (I3RC).
  • Spatial grid - 100?? 100 ?? 36
  • Angular quadrature S30
  • Analytical calculation of un-scattered radiation
    intensity

11
Radiation reflected in zenith
Pixelss optical depth
12
Isotropic conic source. Problem of radiation
safety of astronauts and equipmentin the
spacecraft with the nuclear reactor in motive
regime
Un-scattered component is calculated
semi-analytically Photons field at the
spacecraft top
13
Point source of small aperture.Biomedical problem
  • Un-scattered radiation intensity is defined
    analytically
  • Once-scattered radiation intensity is calculated
    semi-analytically
  • Multi-scattered radiation intensity is determined
    by a grid scheme
  • Spatial grid - 169?100 meshes
  • Angular quadrature S26

Full interior reflecting condition at the top
boundary, refraction coefficient
n1.4. Scattering is simulated by
Heney-Greenstein phase function with asymmetry
parameter g0.8
14
Reflected radiation at the line AB
15
Effectiveness of a parallel algorithm
  • MPI standard
  • Spatial decomposition of calculation region
  • One region ? one processor
  • Effectiveness of a parallel algorithm
  • N is processor number.
  • T(1) is calculation time by one processor.
  • T(N) is calculation time by N processors.

16
We plan to deal with the following manners
  • Peak-forward phase-functions (ray therapy
    problem, atmospheric optic problems)
  • Time-dependent transport equation (impulse source
    in biomedicine problem)
  • Inverse problem (atmospheric optic, biomedicine)
  • These plans rely on experience and intellect and
    energy and optimism of three authors of this
    report, two post-graduate students and two
    students.
  • We are open to cooperate
Write a Comment
User Comments (0)
About PowerShow.com