Re: [femm] FEMM "batch processing"

[dcm3c@xxxxxxx]

In a message dated 4/24/01 3:05:27 PM Eastern Daylight Time, 
frank.e.lenning@xxxxxxxxxx writes:


> Is there a way to "batch process" multiple frequencies in FEMM?  I've 
> been doing simple lightning (F domain) current distribution problems 
> in FEMM but I have to laboriously increment through the frequencies to 
> build a broadband result. (BTW, it works great!) I am interested in 
> getting some quantity (say integral of Ht around a loop) as a function 
> of frequency.  Any help?  I suppose it is possible to do by modifying 
> the code but that sounds painful with no prior experience with this 
> code and limited C++ programming experience. 


I'm glad that the program is proving to be useful to you.  Unfortunately, 
there currently isn't an easy way to do batch processing of this type.  I've 
been threatening to try to implement some sort of scripting for a while, but 
I still haven't decided 

infinitenergy@xxxxxxxxxxx writes:

> Isn't it possible to invent a "shell" for FEMM, which will iteratively run 
> fkern, each time after changing some variable(s) in the model? Then, the 
> changing parameters could be selected from a list of all properties 
> (frequency, materials, magnetization direction, point coordinates, you name 
> it) and the progression of values given either by a list (100, 200, 300...) 
> or a parsed formula (i.e., x = x + 100). This is one suggestion out of many 
> ways it can be done.


I wouldn't want to make the batch run options too restrictive. I've been 
thinking about trying use some existing scripting language to implement batch 
run capabilities (Python?).  One problem is that I haven't really learned how 
to code in any scripting languages yet--oh, well.  Maybe it would be better 
to try to tie it into a program like Scilab (sort of like Matlab, but without 
the cost), because of the built-in plotting and matrix manipulation 
abilities.  I'm a Mathematica junkie, so it might be useful to get the femm 
to talk to Mathematica through mathlink or something.  However, probably most 
of the femm users out there aren't also Mathematica users, so this might not 
be generally that useful.

Hang had suggested to me that it would be good to convert the routines into a 
library 
that can be called from other programs.  This would probably be the easiest 
way to go, would allow some pretty elaborate scripting, and probably be 
pretty useful to me.  However, when it comes right down to it, I'm not sure 
that there are that many user out there who know C++ well enough to take take 
advantage of something like that.

Any more suggestions?

Dave Meeker
--
http://members.aol.com/_ht_a/dcm3c