Statepoint file loading refactor and CAPI function (#2886)

Co-authored-by: Paul Romano <paul.k.romano@gmail.com>
This commit is contained in:
Patrick Shriwise 2024-04-24 17:09:56 -05:00 committed by GitHub
parent 3370ce1978
commit f543c007a3
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 104 additions and 24 deletions

View file

@ -628,3 +628,37 @@ instance, whereas the :meth:`openmc.Track.filter` method returns a new
.. code-block:: sh
openmc-track-combine tracks_p*.h5 --out tracks.h5
-----------------------
Restarting a Simulation
-----------------------
OpenMC can be run in a mode where it reads in a statepoint file and continues a
simulation from the ending point of the statepoint file. A restart simulation
can be performed by passing the path to the statepoint file to the OpenMC
executable:
.. code-block:: sh
openmc -r statepoint.100.h5
From the Python API, the `restart_file` argument provides the same behavior:
.. code-block:: python
openmc.run(restart_file='statepoint.100.h5')
or if using the :class:`~openmc.Model` class:
.. code-block:: python
model.run(restart_file='statepoint.100.h5')
The restart simulation will execute until the number of batches specified in the
:class:`~openmc.Settings` object on a model (or in the :ref:`settings XML file
<io_settings>`) is satisfied. Note that if the number of batches in the
statepoint file is the same as that specified in the settings object (i.e., if
the inputs were not modified before the restart run), no particles will be
transported and OpenMC will exit immediately.
.. note:: A statepoint file must match the input model to be successfully used in a restart simulation.