[Wien] Energy vs Volume is linear!

[pieper]

Hello Marzieh,

I took the freedomn to use your data to test a little fitting procedure 
I just wrote. I tried to fit parameters a, b, c of the function
y=a(x-b)**2+c
to your data.

The fit crashed - probably like the one you may have used - when 
presented the data and asked to fit the parabola from scratch. The 
reason for such a behaviour of a fit procedure frequently is bad user 
behaviour: The program is quite understandably unable to guess 
reasonable starting values for the fit parameters from such data.

Presented with starting values
a=1.000000e-06,   b=9.500000e+03,  c=-8.857000e+04

the fit finds without complaining a very good fit with
a=5.687893e-07,  b=9.612186e+03,  c=-8.856770e+04

So you should do your DFT volume optimization around a volume of 9612 - 
and probably give a little thought to the problem of why you started so 
far off the optimum at larger volumes.

Good luck

Martin Pieper


---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 09.09.2015 03:43, schrieb Marzieh Gh:
> Dear Prof.Blaha & Tran
> 
> I have calculated optimization of supercell (1*1*2). But Energy vs
> Volume is linear!
> 
> case.outputeos:
> 
>    10575.3364   -88567.151612   
>     11162.8553   -88566.366642    
>     12925.4112   -88561.471064   
>     11750.3745   -88565.095045   
>     12337.8929   -88563.437182    
> 
> Is this normal?  If no what is reason? What do I do?
> 
> Please help me
> 
> Best Regards
> 
> --
> 
> Marzieh Ghoohestani
> PhD Student of Computational Nano Physics
> Nano Research Center, Department of Physics
> University of Technology, Isfahan, Iran
> 
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html