PyTeCK¶
PyTeCK (Python-based Testing of Chemical Kinetics) automatically evaluates the performance of chemical kinetic models against experimental data. It reads experimental ignition-delay measurements in the ChemKED YAML format, simulates the corresponding cases with Cantera, and reports how well a chemical kinetic model reproduces the measurements.
PyTeCK is released under the MIT license and developed openly on GitHub.
Features¶
Evaluate a kinetic model against whole datasets of experimental ignition-delay data described with ChemKED.
Support for shock tube and rapid compression machine (RCM) experiments, including non-ideal effects such as pressure rise and volume-time histories.
Automatic, per-case parallel execution of simulations.
A quantitative error function summarizing model performance across a dataset.
Quick start¶
Install from PyPI (see Installation for conda and development installs):
$ pip install pyteck
Run an evaluation from the command line:
$ pyteck --model mech.yaml --model-keys spec_keys.yaml --dataset dataset_file.txt
or from Python:
from pyteck.eval_model import evaluate_model
output = evaluate_model(
model_name="mech.yaml",
spec_keys_file="spec_keys.yaml",
dataset_file="dataset_file.txt",
data_path="data",
model_path="models",
results_path="results",
)
print(output["average error function"])
The Example: hydrogen shock tube page walks through a complete, runnable hydrogen shock-tube
case. The full list of command-line options is available with pyteck --help.
User guide¶
API reference¶
Citation¶
If you use PyTeCK in a scholarly publication, please cite it as described in CITATION.md.
