Python wrapper for batch simulations (e.g. Cantera).
Warning
This repository is archived and no longer maintained. It is in a read-only state. No further updates, issues, or pull requests will be accepted. Interested users are encouraged to either fork the repository to continue development independently, or submit a feature request to Cantera/enhancements for similar capabilities. In either case, limited support from the original developer may be available.
Sphinx documentation includes:
- A brief summary (see Overview),
- Typical batch simulation descriptions (see Batch Jobs), and
- Jupyter Notebooks with illustrated examples.
The software was developed with the following objectives in mind. It should:
- provide a low-level interface to essential
canteracapabilities - create a flexible framework for generic simulations
- provide a command line interface
- be easily scriptable (using YAML configuration files)
- enforce units (via
pint) - enable parallel execution of (single-threaded) simulations (via
multiprocessing) - enable simple re-import of simulation results into native Cantera objects
Although core functions of this software are continuously tested, there may be remaining bugs, non-working features, or other issues that could prevent a user from using this software to their specification. If you find problems, please report them in the issue tracker.
The simplest approach is to install ctwrap via pip, i.e.
$ pip install ctwrap
Note that
ctwrapitself does not depend on a Cantera installation (only simulation modules do). It is recommended to install Cantera before ctwrap in a separate step.
A parallel batch job for adiabatic flame calculations uses the freeflame module (modified from Cantera's adiabatic_flame.py example). Based on the YAML configuration given as a reference, a variation of 12 equivalence ratio values is run as:
$ ctwrap run freeflame freeflame.yaml --parallel --strategy sequence
Results are written to a single file freeflame.h5.