PDE2D

Definition

PDE2D is a finite element code developed by Granville Sewell at UT El Paso and UT System personnel that solves systems of linear and nonlinear time-dependent, steady-state, and eigenvalue partial differential equations in general two-dimensional regions and in three-dimensional boxes. PDE2D can handle problems in such areas as elasticity, diffusion, heat conduction, potential energy, and fluid mechanics. It has an interactive interface, that makes it extremely easy-to-use, and extensive graphical output capabilities.

Usage

Note: a sample UNIX Bourne script "pde2d" has been provided for step 1, and the script "runpde2d" has been provided for steps 3-4.
  1. Execute "pde2d.pre".
    After answering all the interactive questions, a main program "main.f" will have been created.
  2. Edit "main.f".
    if necessary, to make minor corrections or modifications.
  3. Compile "main.f", using a Fortran77 compiler, and link the object code with the "libpde2d.a" library and your local GKS library.
    If you made any of the optional substitutions discussed in step A4 above, you may need to also load the appropriate library (CalComp or BLAS).
  4. Execute the resulting program.
    Tabular output (if any) will be written to the file "pde2d.out"
  5. Plots
    When a PDE2D program is run, a CGM metafile "meta.cgm" will be created; X-window plots can be made from this metafile using the command

    mi meta.cgm

    To page to each new plot you have to press the "Break" key. If your keyboard doesn't have a Break key, you can set one in your .xinitrc file, for example:

    xmodmap -e 'keysym F1 = Break'

    sets the F1 key to be the Break key.
  6. To convert "meta.cgm" to a color PostScript file "postfile", type:

    gplot -dPSC meta.cgm postfile

  7. The user can see example files on /usr/local/lib/pde2d/examples