SPINS User Guide: Difference between revisions
No edit summary |
No edit summary |
||
Line 55: | Line 55: | ||
FILE * en_record = fopen("energy_record.txt","a"); | FILE * en_record = fopen("energy_record.txt","a"); | ||
assert(en_record); | assert(en_record); | ||
fprintf(en_record,"%.8f | fprintf(en_record,"%.8f %.14g\n",time,ke); | ||
fclose(en_record); | fclose(en_record); | ||
} | } |
Revision as of 09:59, 12 July 2012
Welcome to the SPINS case setup page.
Getting SPINS and compiling
(Chris: Can we put it on GitHub or make a new repo?)
SPINS consists of a bunch of C++ source files and a bunch of case files, and it requires four libraries. UMFPack, AMD and Blitz++ are supplied with SPINS, and it uses the system-installed FFTW.
Directory structure:
- cpp/src - SPINS source files
- cpp/src/cases - a few dozen example case files
- cpp/AMD - AMD library
- cpp/UMFPack - UMFPack library
- cpp/UFconfig - Helper for compiling AMD/UMFPack
To compile, first ....
The basics
The SPINS model is a Navier-Stokes solver that gets parameters and initial/boundary conditions from calls to user-provided routines. The user-provided routines are encapsulated in the class BaseCase (see BaseCase.hpp).
Creating your own custom configuration involves building a derived class based on BaseCase. The case file cases/doc_minimal.cpp shows the structure of a case file. It makes sense to start with another similar case file and customise it.
Examples of common operations
Outputting the grid
At the top of your file include the global tensor indices:
// Tensor variables for indexing
blitz::firstIndex ii;
blitz::secondIndex jj;
blitz::thirdIndex kk;
Create array tmp and generate the grid, then write the array and the grid reader out:
tmp = xx(ii) + 0*jj + 0*kk;
write_array(tmp,"xgrid");
write_reader(tmp,"xgrid",false);
tmp = 0*ii + yy(jj) + 0*kk;
write_array(tmp,"ygrid");
write_reader(tmp,"ygrid",false);
tmp = 0*ii + 0*jj + zz(kk);
write_array(tmp,"zgrid");
write_reader(tmp,"zgrid",false);
Energy diagnostic
If you're on a periodic grid, use this for kinetic energy diagnostic
double ke = 0.5*rho_0*pssum(sum(u*u+v*v+w*w))/(Nx*Ny*Nz)*Lx*Ly*Lz; // KE
if (master()) {
FILE * en_record = fopen("energy_record.txt","a");
assert(en_record);
fprintf(en_record,"%.8f %.14g\n",time,ke);
fclose(en_record);
}
and if you're on a mapped chebyshev grid, you can use this for the KE computation
double ke = pssum(sum((*get_quad_x())(ii)*(*get_quad_y())(jj)*(*get_quad_z())(kk)*
(pow(u(ii,jj,kk),2)+pow(v(ii,jj,kk),2)+pow(w(ii,jj,kk),2))));
Initialising velocities
ToDo: More here!!