Impossible to perform peridynamics simulation from a data file

Hi all,

I want to perform a peridynamics simulation from a data file.
But I have this error: ERROR: Pair peri requires a lattice be defined (pair_peri_pmb.cpp:366)

Why have you taken a such precaution ?

Is it possible to overcome this limitation without changing the code ?

Thanks in adavance for your reply
JF

Hi all,

  I want to perform a peridynamics simulation from a data file.
But I have this error: ERROR: Pair peri requires a lattice be defined
(pair_peri_pmb.cpp:366)

Why have you taken a such precaution ?

one would have to take a closer look at the code.

Is it possible to overcome this limitation without changing the code ?

yes. you can always define a lattice. it is just
a convenient way to compute coordinates on
a regular grid. it usually has no physical impact
on the simulation itself and usually independent
from using a data file. depending on what the lattice
gridpoints are used for, it may need to be adjusted
to match the contents of the data file.

axel.

Hi all,

  I want to perform a peridynamics simulation from a data file.
But I have this error: ERROR: Pair peri requires a lattice be defined
(pair_peri_pmb.cpp:366)

Why have you taken a such precaution ?

please have a look at:
http://www.sandia.gov/~mlparks/papers/PDLAMMPS.pdf

it clearly says that only a primitive cubic lattice is supported
and that you *have* to include a lattice command in
your input script.

from looking at the code, it is evident that the grid
defined in the lattice is integral part of the implementation
of the algorithm and that it was done under the assumption
that the lattice has the same length vectors in x, y, and z.

please also read on in the PDF how the lattice affects
other simulation parameters.

Is it possible to overcome this limitation without changing the code ?

see my previous e-mail on how to circumvent the check,
but it is up to you, to make sure your input data is consistent
with the requirements on the implementation. otherwise,
you *will* have to do some (probably significant) programming
or risk computing bogus data.

axel.

Hi Axel,

Thanks for your quick replies.
Now I understand that I have to define a lattice constant, because this constant is used in the pair_peri_*.cpp files.

I will watch out for loading a data file with the points which respect this lattice constant.

Best regards
JF