archive-edu.com » EDU » R » RPI.EDU

Total: 920

Choose link from "Titles, links and description words view":

Or switch to "Titles and links view".
  • RPI SCOREC - Technical Reports
    REPORTS SOFTWARE FACILITIES EMAIL SERVICES WIKIS Author D C Haworth K Jansen Title Large eddy simulation on unstructured deforming meshes towards reciprocating IC engines Year 2000 Journal Computers Fluids Volume

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=21 (2015-07-15)
    Open archived version from archive


  • RPI SCOREC - Technical Reports
    IJCFD Volume 9 Pages 227 235 Abstract The purpose of this paper is to develop a consistent method to calculate the fluxes or the stresses in an incompressible flow particulary a turbulent flow A post processing technique has been developed such that fluxes or stresses are treated as unknown values which are solved for weakly using the finite element method The method has been applied to calculate the Reynolds stresses

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=22 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    the variational multiscale method Year 2000 Journal Computing and Visualization in Science Volume 3 Pages 47 59 Abstract A Large Eddy Simulation LES formulation is developed from the variational miltiscale method Modeling is confined to the effect of a small scale Reynolds stress in contrast with classical LES in whitch the entire subgridscale stress is medeled All other effects are accounted for exactly Is is argued that many shortcomings of

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=23 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    parallel computers Algorithms are presented for h adaptative simulations for transient fluid calculations which rely on error indicators based on statistics of various flow quantities This enables the construction of adaptively refined meshes based on statistically steady error indicators rather than the traditional steady quantities To make effective use of these advanced simulation procedures sugnificant effort is required to design and develop advanced software This paper presents implementational details pertaining

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=24 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    dynamics simulations using parallel computers Year 2000 Abstract The preprocessing stage of finite element analysis of the Navier Stokes equations is becoming increasingly important as the desire for more general boundary conditions as well as applications to parallel computers increases The set up of general boundary conditions and communication structures for parallel computations should be accomplished during the preprocessing phase of the analysis if possible to ensure efficient computations for

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=25 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    method to complex problems involving h and p refinment and non positive definite systems We extend the discussion of the preconditionner by considering an alternative to the algebraic coarsenong This spacial coarsener leverages the TRELLIS octree structure to impose regular coarsening on the irregular mesh comprising the discrete problem domain The octree structure is defined and the operations required for the serial generation of the octree coarsened system are explained We demonstrate that this coarsener gives superior convergence behavior as the problem size grows via h refinment We also consider one of the available submatrix approximations required for the AMLI preconditionner and derive an alternate approximation to Symmetric Successive Over Relaxation SSOR which can be computed with a single SOR like iteration We prove that under certain conditions of diagonal dominance this approximation to SSOR is valid and show that it works similary to SSOR in practice by applying it to several test problems The second effort is concerned with an efficient and maintenable parallel implementation of the algorithm We present an abstraction for the matrix and vector objects sufficient for the implementation of Krylov space techniques and apply that abstraction to the generation of a transparently parallel generic library

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=27 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    M S Shephard Title AN ADAPTIVE DISCONTINUOUS GALERKIN TECHNIQUE WITH AN Year 2000 Journal Siam Abstract We present a high order formulation for solving hyperbolic conservation laws using the Discontinuous Galerkin Method DGM We introduce an orthogonal basis for the spatial discretization and use explicit Runge Kutta time discretization Some results of higher order adaptive re nement calculations are presented for inviscid Rayleigh Taylor ow instability and shock re exion

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=28 (2015-07-15)
    Open archived version from archive

  • RPI SCOREC - Technical Reports
    reference element The edge coordinate functions of a reference element are allowed to be either nodal with special choices of nodes or hierarchical polynomials of several types In relation to the definition of these coordinate functions within the elements we also distinguish two cases arbitrary and so called discrete quasi harmonic coordinate functions The latter are obtained by means of explicitly given and nonexpensive prolongation operators In all these situations

    Original URL path: http://www.scorec.rpi.edu/reports/view_report.php?id=56 (2015-07-15)
    Open archived version from archive



  •