search

ORNL-Fusion / ips-wrappers

IPS wrapper / helper codes.
9 stars 4 forks source link

TSC with fastran_init #16

Open dlg0 opened 9 years ago

dlg0 commented 9 years ago

@batchelordb

I'm trying to do a TSC based simulation using fastran_init to create the initial plasma state from a set of in* files. My present sticking point is that the IPS copies the plasma state into the TSC work dir, then TSC produces a new plasma state that has 26 species in it (there should not be that many), i.e., ...

my run is in 

/project/projectdirs/atom/users/greendl1/diem_tsc_error11
greendl1@edison02:/project/projectdirs/atom/users/greendl1/diem_tsc_error11/work/epa__tsc_4> ncdump state-0.nc | grep dim_nspec_th
    dim_nspec_th = 3 ;
    dim_nspec_tha = 3 ;
greendl1@edison02:/project/projectdirs/atom/users/greendl1/diem_tsc_error11/work/epa__tsc_4> ncdump ss31615_3_ps.nc | grep dim_nspec_th
    dim_nspec_th = 25 ;
    dim_nspec_tha = 5 ;

The state-0.nc file is the state prior to running TSC init, and the ss31615_3_ps.nc is after.

I'm off to dig into the TSC driver, but any tips here would be great.

Thanks.

parkjm commented 9 years ago

@dlg0 Were you able to accomplish this task? I'm sure that it's not easy and involves many housekeepings etc. If we can do this, however, it means, more or less, that we can use a TSC component as an equilibrium solver component in general for a variety of workflows we are working on.

dlg0 commented 9 years ago

@parkjm @diemsj @batchelordb

So last night I did some hacking on pstool, ips-fastran/src/genray.py and ips-fasran/src/zplasmastate.py to make it possible to run the TSC workflow but using the ips-fastran/src/genray.py wrapper instead of the CSWIM genray wrapper.

As you point out, it's probably a bit too much effort to get TSC to work with fastran_init, but I think the primary issue with the TSC / FASTRAN comparison is getting the source components to work from the same input files.

So, looking at the existing ips-fastran/src/genray.py the reason it wouldn't work with the TSC initialized plasma state was because it expected all it's plasma state initializations to be done by fastran_init, rather than in the init section of ips-fastran/src/genray.py. So all I've done is add that. It seems to be working even. I'll push these changes to the dlg branch of ips-fastran so you can review them.