This project implements the BOCOPSolution
Python class, which reads a BOCOP solution directory for ease of use in scientific and engineering applications.
For example, it implements:
- Grouping states/adjoint states/control variables together in
VariableBunch
classes. - Representing variables in common types for scientific Python, such as NumPy
ndarray
and PandasDataFrame
. - Improved plotting with the matplotlib library, easily plotting variables over time as well as phase diagrams between states with full customization capabilities.
- Variable interpolation with cubic splines, allowing for easy differentiation and integration.
- Variable interpolation with piecewise constants, numerically detecting bang-bang controls and returning their LaTeX representation.
For local or Google Colab installation, just use the command:
pip install git+https://github.com/johnny-godoy/bocop-reader.git
The main object that stores all solution files is the BOCOPSolution
class, which can be imported like such:
from bocop_solution import BOCOPSolution
We include a small guide in the notebooks/Introduction.ipynb
notebook.
It is also available as HTML here.
For bug reports and user help, I recommend you to use the GitHub Issues feature, but you may also contact the author at [email protected].